UNIVERSITY OF GHANA COLLEGE OF HEALTH SCIENCES SCHOOL OF PUBLIC HEALTH PRE-PACKAGED FOOD LABELLING AND USE AMONG HOUSEHOLDS IN ACCRA GEOFFREY ADEBAYO ASALU (10085710) A THESIS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF DOCTOR OF PHILOSOPHY PUBLIC HEALTH DEGREE DEPARTMENT OF POPULATION FAMILY AND REPRODUCTIVE HEALTH SEPTEMBER 2021 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I hereby declare that this thesis ‘Pre-packaged food labelling and use among households in Accra’ is a product of my original independent research work. Professor Richmond Aryeetey was my lead supervisor and other members of the supervision team were Dr Adom Manu and Professor Amos Laar. I affirm that this is my original work and it has not been submitted for award of any degree or examination. All references used or quoted have been duly acknowledged. ………………………. …………………………. Geoffrey Adebayo Asalu Date PhD Candidate 23/08/2022 ………………………… ………………………….. Professor Richmond Aryeetey Date (Principal Supervisor) 25 August 2022 ……………………….. ………………………….. Dr. Adom Manu Date (Supervisor) 29 August 2022 ………………………….. …………………………… Professor Amos Laar Date (Supervisor) 23/08/2022 University of Ghana http://ugspace.ug.edu.gh ii DEDICATION I dedicate this work to the glory of our Father Lord Jesus Christ for his unfailing love and grace. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGMENTS The inspiration, the progress and the completion of this PhD journey were borne out of the contributions of many that I am forever grateful for. In the first place, I would like to thank God for the grace, strength, and wisdom to complete this work. I cannot imagine how I could have undertaken this PhD work without the support and encouragement of my lead supervisor Professor Richmond Aryeetey. Richmond has been God-sent. His mentorship style is exceptional. I underwent professional training that will forever change the course and direction of my career. I appreciate his generous mentorship and the many opportunities he offered me. I am indebted to the other members of my supervisory team Dr. Adom Manu and Professor Amos Laar for their valuable contributions to this work. To the rest of the faculty members of the Department of Population, Family, and Reproductive Health, School of Public Health, the 2016 cohort PhD candidates, Prof Aryeetey’s lab colleagues I am grateful. I would like to thank my many financial funders including GET-Fund and family and friends especially my dad Moses Asalu and Ayodele Asalu my brother. Others include Anthony Odiyi, Leye Falade and Kofi Akyea. I would also like to express my gratitude to the School of Public Health, UHAS especially the department of Family and Community Health for their support. I am indebted to Wisdom Axame who was always supportive in my data analysis. To my precious jewel and sweetheart Bubune for her support, especially reviewing all my write-ups. I thank the whole Asalu family and friends for their prayers and encouragement to see this work through. University of Ghana http://ugspace.ug.edu.gh iv ABSTRACT The rate of diet-related non-communicable diseases (NCDs) is increasing rapidly in Africa. Consumption of pre-packaged foods which are nutrient-poor and calorie-rich is one of the main drivers of NCDs. Food labels can guide consumers to make healthy food choices. In Ghana, there is paucity of evidence on the type of nutrition and health- related label information that consumers are exposed to, and whether consumers use such information in their purchase decisions. Therefore, this study assessed label characteristics of pre-packaged foods, and determined the drivers of label use among consumers in urban Accra. This study employed a quantitative cross-sectional multi-method design. A survey of food labels was conducted in community-based retail shops. Information on labels of pre-packaged foods were obtained by taking pictures and analysing the contents based on the International Network for Food and Obesity/Non-Communicable Diseases Research, Monitoring and Action Support’s (INFORMAS) taxonomy. Besides, respondents (510) were selected using a multi-stage sampling technique. Participants were interviewed, using a structured pre-tested questionnaire. Questions assessed consumers’ pre-packaged food use behaviour, their perceptions of health-related label information, understanding of food labels and socio-demographic covariates. Three hundred and fifty-one (N=351) pre-packaged foods were sampled. Out of 343 labelled products, 68.8% had nutrition declaration information. Back- of-Pack (BOP) nutrition label formats were dominant (87.3%) compared to the Front-of -Pack (FOP) format. Guideline Daily Amount (GDA) was the commonest FOP identified. Nutrition claims were twice as frequent as health claims. Most products (>84%) with health and nutrition claims complied with FDA and Codex Alimentarius standards. Although the disclosure of nutrition and health-related information was appreciable it did not meet University of Ghana http://ugspace.ug.edu.gh v the current recommendations set out by Codex and INFORMAS in promoting a healthy environment. Majority of survey respondents (79.4%) were females and had at least secondary school-level education (65.3%). Most households (77%) purchased pre-packaged foods from traditional markets or corner shops and a few (7%) purchased from supermarkets. Most consumers had a positive perception of food labels, and their purchase behaviour was driven more by taste and price. Although most (62%) respondents reported adequate understanding of labels, objective assessments showed a lower (9%) consumer knowledge and skill in using labels. Only a third of respondents were food label users. Labels were used mainly to ascertain product safety and less likely for nutrition and health reasons. Higher proportions (45%) of non-label users indicated technical nature of label information and limited understanding as reasons for not using labels. In multiple logistic regression modelling, being part of a larger household (AOR: 2.41; 95% CI: 1.16-4.99), having tertiary education (AOR: 6.75; 95% CI: 1.99-22.88), adequate nutrition-related knowledge of labels (AOR: 1.92; 95% CI: 1.15-3.20) and food label understanding (AOR: 2.51; 95% CI: 1.19-5.29) as well as individuals with self-reported understanding of labels (AOR: 10.06; 95% CI: 2.40-42.27) predicted food label use. However, a multiple linear regression analysis the following variables: levels of education, individuals with food allergies, those who had a previous education on the use of labels, those who perceived labels are easy to understand and individuals who self-reported adequate food label understanding showed positive association with the use of health-related label information. Therefore, these findings suggest that educational interventions and labelling policy reforms are needed to encourage, enable, and improve consumer use of nutrition and health-related information on food labels in Ghana. University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENTS DECLARATION ............................................................................................................ i DEDICATION ............................................................................................................... ii ACKNOWLEDGMENT.............................................................................................. iii ABSTRACT .................................................................................................................. iv TABLE OF CONTENTS .............................................................................................. vi LIST OF TABLES ........................................................................................................ ix LIST OF FIGURES ....................................................................................................... x LIST OF ABBREVIATIONS ....................................................................................... xi CHAPTER ONE ............................................................................................................ 1 1.0 INTRODUCTION ................................................................................................... 1 1.1 Background .......................................................................................................... 1 1.2 Statement of the Problem ..................................................................................... 4 1.3 Justification .......................................................................................................... 7 1.4 Objectives of Study .............................................................................................. 8 1.4.1 General Objective .......................................................................................... 8 1.4.2 Specific objectives ......................................................................................... 8 CHAPTER TWO ........................................................................................................... 9 2.0 LITERATURE REVIEW ........................................................................................ 9 2.1 Food processing and unhealthy diets ................................................................... 9 2.2 The burden of diet-related NCDs ....................................................................... 13 2.3 Food Labelling Policies addressing unhealthy diets and the obesity epidemic . 14 2.4 Food and nutrition labelling of Pre-packaged Foods and their Regulation........ 18 2.4.1 Food Labelling of pre-packaged foods a standard of Codex ....................... 18 2.4.2 Nutrition and Health Information on Pre-packaged Foods ......................... 19 2.4.3 Voluntary and Mandatory Nutrition Labelling of Pre-packaged foods ....... 29 2.5 Consumer food label use .................................................................................... 33 2.6 Reasons for use and non-use of nutrition information on food labels ............... 35 2.7 Consumer perception and understanding of food labels .................................... 36 2.8 Factors associated with Consumer food label use .............................................. 37 2.9 Conceptual Framework ...................................................................................... 39 2.10 Limitation of the Literature Review ................................................................. 43 CHAPTER THREE ..................................................................................................... 44 3.0 METHODOLOGY ................................................................................................ 44 3.1 Introduction ........................................................................................................ 44 University of Ghana http://ugspace.ug.edu.gh vii 3.2 Philosophical Basis of the Study ........................................................................ 44 3.2 Research Design ................................................................................................. 45 3.4 Study Location ................................................................................................... 46 3.5 Pre-packaged Food label Survey ........................................................................ 50 3.5.1 Sampling Locations ..................................................................................... 50 3.5.2 Sampling of Food Retail Outlets ................................................................. 50 3.5.3 Sampling size for Food Products ................................................................. 50 3.5.4 Pre-packaged Food Products: Inclusion/Exclusion Criteria ........................ 50 3.5.5 Data Collection procedure ........................................................................... 51 3.5.6 Data Entry, Processing and Management .................................................... 52 3.5.7 Label Characteristics (Variables) Measured ................................................ 53 3.5.8 Quantitative Content Analysis ..................................................................... 56 3.6 Consumer Household Survey ............................................................................. 57 3.6.1 Sampling ...................................................................................................... 57 3.6.2 Inclusion/ Exclusion criteria ........................................................................ 60 3.6.3 Data Collection for Consumer Household Survey ...................................... 61 3.6.4 Survey Variables and measures ................................................................... 63 3.6.5 Data Analysis for Consumer Survey ........................................................... 68 3.7 Quality Control/Assurance ................................................................................. 71 3.7.1 Training of data collectors ........................................................................... 71 3.7.2 Ethical Considerations for the Study ........................................................... 71 CHAPTER FOUR ........................................................................................................ 72 RESULTS .................................................................................................................... 72 4.0 Introduction ........................................................................................................ 72 4.1 Content Analysis of Food Labels ....................................................................... 72 4.1.1 Overview of Pre-packaged Food Products sampled ....................................... 72 4.1.2 Nutrition and Health-related Label Information on Pre-packaged Foods ... 73 4.1.3 Nutrition label information .......................................................................... 74 4.1.4 Food label claims across various pre-packaged foods ................................. 76 4.1.5 Health and nutrition claims .......................................................................... 77 4.1.6 Compliance of food labels to labelling standards. ....................................... 78 4.2 Consumer Survey ............................................................................................... 81 4.2.1 Household and Socio-demographic characteristics of respondents. ........... 81 4.2.2 Usual places for purchasing pre-packaged foods. ....................................... 83 4.2.3 Consumers’ considerations while purchasing pre-packaged foods. ............ 83 University of Ghana http://ugspace.ug.edu.gh viii 4.2.4 Household Pre-packaged food dietary patterns. .......................................... 85 4.2.5 Consumers perceptions towards nutrition and health-related information on pre-packaged foods ............................................................................................... 86 4.2.6 Nutrition-related knowledge of food labels ................................................. 87 4.2.7 Food label understanding of respondents. ................................................... 88 4.2.8 Factors associated with respondents’ nutrition-related knowledge of food labels. .................................................................................................................... 90 4.2.9 Factors associated with food label understanding of respondents ............... 93 4.2.10.0 Food label usage of respondents ............................................................. 95 4.2.11 Factors associated with food label use ...................................................... 98 4.2.12 Predictors of health-related information use (HIU) among food label users ............................................................................................................................ 101 CHAPTER FIVE ....................................................................................................... 103 5.0 DISCUSSION ...................................................................................................... 103 5.1 Introduction ...................................................................................................... 103 5.2 The disclosure of nutrition and health-related Information on pre-packaged food products .................................................................................................................. 103 5.3 Compliance of Food labels and Nutrition/Health claims to FDA and Codex Requirements .......................................................................................................... 109 5.4 Pre-packaged food use behaviour among households in Accra ....................... 113 5.5 Consumer perceptions, nutrition knowledge and food label understanding among consumers in Accra .................................................................................... 115 5.5 Food label use behaviour, reasons for using and not using food labels ........... 117 5.6 Factors associated with consumer understanding and use of health-related label information on pre-packaged foods........................................................................ 120 5.7 Strengths and contributions to knowledge ....................................................... 123 5.8 Limitations of the study.................................................................................... 124 CHAPTER SIX .......................................................................................................... 126 CONCLUSION AND RECOMMENDATIONS ...................................................... 126 6.1 Conclusions ...................................................................................................... 126 6.2 Recommendations ............................................................................................ 128 REFERENCES .......................................................................................................... 130 APPENDICES ........................................................................................................... 149 Appendix A: Data Collection Instruments ............................................................. 149 Appendix B: Consent Form and Ethical Approval Letter ...................................... 178 Appendix C: Supplementary Results for Chapter Four ......................................... 182 University of Ghana http://ugspace.ug.edu.gh ix LIST OF TABLES Table 1. Classification of Communities in Accra by Socio-economic status (SES). .. 48 Table 2: Summary of variables and their form of utilization in the survey’s data analysis ........................................................................................................... 68 Table 3. Health-related and Nutrition label information on sampled Pre-packaged Foods obtained from neighborhood shops in Accra. ...................................... 73 Table 4: Types of Health and Nutrition Claims and format of presentation on Labels ........................................................................................................................ 78 Table 5: Compliance of food labels to Basic Labelling Requirements by the Food and Drugs Authority, Ghana (N=343) .................................................................. 79 Table 6: Household and Socio-demographic characteristics of survey respondents (N=510) .......................................................................................................... 82 Table 7: Consumers perception towards nutrition and health-related information on pre-packaged foods ......................................................................................... 87 Table 8: Respondents’ levels of understanding of health-related food label information (n=510) ....................................................................................... 90 Table 9: Factors associated with respondents’ nutrition-related knowledge of food labels. .............................................................................................................. 92 Table 10: Factors associated with respondents’ understanding of food labels ............ 94 Table 11: Consumers’ reasons for non-use of pre-packaged food labels (N=348) ..... 98 Table 12: Factors associated with food label use......................................................... 99 Table 13: Predictors of health-related use (HIU) indexa among food label users(N=162) ................................................................................................ 102 University of Ghana http://ugspace.ug.edu.gh x LIST OF FIGURES Figure 1: INFORMAS Taxonomy for describing health-related aspect of food labelling .......................................................................................................... 6 Figure 2: Examples of Nutrition label BOP adapted from(Campos et al., 2011) ........ 22 Figure 3: Examples of FOP labelling systems adapted from Chantal, J., Hercberg, S., & World Health Organization. (2017). ......................................................... 27 Figure 4: Timeline of countries all over the world adopting FOP labeling system. .... 32 Figure 5: An adapted Conceptual Framework of consumer understanding and use of food label ...................................................................................................... 42 Figure 6: Map of Accra Metropolitan Area (Source; Ghana Statistical Service) ........ 49 Figure 7: Schematic Diagram of Sampling Design ..................................................... 60 Figure 8: Proportion of Pre-packaged Food Products categories sampled from neighbourhood shops in Accra...................................................................... 72 Figure 9: Back of Pack (BOP) Nutrition labels across various pre-packaged food categories:“Big 4” =energy value, carbohydrates, fat, protein; “Big 8” = “Big 4” plus sugar, saturated fat, sodium and fibre ...................................... 74 Figure 10: Types of FOPS schemes and their country of origin .................................. 75 Figure 11: Types of FOPs on various pre-packaged food categories .......................... 76 Figure 12: Distribution of different types of Health-related claims on various categories of pre-packaged foods according to INFORMAS taxonomy ...... 77 Figure 13: Compliance with FDA Labelling Guidelines (index) for various categories of pre-packaged foods. .................................................................................. 80 Figure 14: Usual places of purchase of pre-packaged foods ....................................... 83 Figure 15: Consumers’ considerations while purchasing pre-packaged foods ............ 84 Figure 16: Consumers’ rankings of most important reasons for choosing pre-packaged food products. ............................................................................................... 85 Figure 17: Household Patterns of Frequently Purchased and Consumed Pre-packaged food Categories. ............................................................................................ 86 Figure 18: Levels of food label nutritional knowledge of Respondents ...................... 88 Figure 19: Levels of understanding of food label among respondents ........................ 89 Figure 20: Food label usage ......................................................................................... 95 Figure 21: Frequency of checking food label components .......................................... 96 Figure 22: Reasons for reading food labels. ................................................................ 97 University of Ghana http://ugspace.ug.edu.gh file:///C:/Users/SPH-EL/Downloads/Thesis_geoffreyasalu1edit_corrections.docx%23_Toc109134218 xi LIST OF ABBREVIATIONS AMA - Accra Metropolitan Area BMI - Body Mass Index BOP- Back- of- Pack CAC - Codex Alimentarius Commission CVD – Cardio-vascular diseases DALYs - Disability-adjusted life years EA - Enumeration Areas FAO - Food and Agriculture Organization FDA - Food and Drugs Authority - FOP - Front-of-Pack GDA - Guideline daily allowance GDA - Guideline Daily Amounts GDHS - Ghana Demographic Health Survey GHS - Ghana Health Service GSS - Ghana Statistical Service HICs - High-income countries HIU - Health-related Information Use HNCs – Health and Nutrition Claims INFORMAS - International Network for Food and Obesity/NCDs Research, Monitoring and Action Support LMICs - Low-to-middle-income countries LMICs, - Low-to-middle income countries MP - Minimally Processed foods MTL - Multiple Traffic Lights NCDs - Non-Communicable Diseases NFT - Nutrition Fact Table NRV - Nutrient Reference Values OECD - Organization for Economic Co-operation and Development PCI - Processed culinary ingredients. PF - Processed foods PHC - Population and Housing Census University of Ghana http://ugspace.ug.edu.gh xii SES - Socioeconomic status UPFDs - Ultra-processed foods and drinks WCRF - World Cancer Research Fund WHO - World Health Organization WL - Warning labels University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE 1.0 INTRODUCTION 1.1 Background Non-Communicable Diseases (NCDs) negatively affect the health, socio- economic well-being, and development of the population (Murray et al., 2020). The prevalence of NCDs is increasing worldwide, especially, in Low- and -Middle-income countries(LMICs) (WHO, 2019). Heart disease, diabetes, cancer, and chronic respiratory disease are the leading cause of premature death and disability in the region (Bigna & Noubiap, 2019). Disability Adjusted Life Years (DALYs) due to NCDs in Africa has risen by 67% from 1990 to 2017 (Gouda et al., 2019). In Ghana, the Global Burden of Disease Report estimated that 40% of all deaths were linked to NCDs (IHME, 2019). Ischemic heart disease, stroke , diabetes and lower respiratory diseases are the major causes of morbidity and mortality in the Ghanaian population (Adu-Gyamfi et al., 2020). This growing burden of NCDs coupled with high prevalence of infectious diseases is causing huge financial loss and putting undue pressure on the already weak healthcare system (Nyaaba et al., 2020). NCDs are rooted in a complex web of causes including genetic, physiological, environmental, and behavioural factors (Michele Cecchini et al., 2010; Wright & Aronne, 2012). However, modifiable lifestyle factors such as smoking, harmful use of alcohol, unhealthy diets, and physical inactivity are the main drivers of NCDs (Hazreen et al., 2014). These lifestyles also contribute, significantly, to increasing rates of obesity and high blood pressure, which are major metabolic risk factors for NCDs. Obesity ranks sixth among the world’s leading causes of disability-adjusted life years (DALYs) (Ford et al., 2017). The majority of the world’s obese people live in developing countries (Ng et al., 2014). University of Ghana http://ugspace.ug.edu.gh 2 The obesity epidemic has been linked with changes in diet as a result of increasing income, rapid urbanization, and advancement in technology(Hawkes et al., 2017). These diet-related transitions are characterized by consumption of energy-dense foods such as fast foods, and ultra-processed foods which are high in fat, salt, and sugar (Bielemann et al., 2015; Carlos A Monteiro & Cannon, 2012). Consumption of these foods is motivated by increased availability and access, increased household affordability, convenience and palatability (Hawkes et al., 2017). The changes in eating habits, from traditional diets (typically made of whole foods such as legumes, cereals and whole grains, and low fat, salt and sugar) to diets high in salt, refined sugar, and oils are typical of the Nutrition transition ( Hawkes et al., 2017; Popkin et al., 2011). The World Health Organization (WHO) has proposed population- and individual-level recommendations for reducing the burden of obesity and other diet- related NCDs (WHO, 2013). The recommendations include (i) limiting high energy intake from total fats especially from saturated fat, (ii) eliminating trans-fat, (iii) limiting intake of free sugars and (iv) limiting salt intake from all sources. A supportive environment that enhances healthy food choices is important for improved nutrition. Consumer awareness and use of labels on pre-packaged foods and beverages constitute a potentially cost-effective means for addressing the rising burden of obesity and diet-related NCDs (Michele Cecchini et al., 2010; Corvalán et al., 2013; Hawkes et al., 2015). Food labels represent the interface of communication between consumers of pre-packaged foods and manufacturers. Labels give information on food composition, ingredients and their relative amounts, nutrient content, origin of the product, and preservation (Rayner et al., 2013). Therefore, labels can serve as important public University of Ghana http://ugspace.ug.edu.gh 3 health tools to help consumers make informed decisions when purchasing or using pre- packaged foods. On the other hand, food labels have the potential of misinforming and misleading consumers in their food choice, especially information related to health and nutrition claims (Hawkes et al., 2013;Hawkes, 2004). Although food and nutrition claims can provide consumers with information on potential nutrition and health benefits of products, if such benefits have been substantiated, there is debate on how useful they are considering that the primary purpose for their use by manufacturers is for marketing rather than to aid consumers to select healthy foods (Bialkova et al., 2016). To this end, the Codex Alimentarius Commission has developed standards and guidelines for nutrition labelling on food products (Codex, 2001). Many nations have adapted these guidelines; or aspects of them in labelling regulations, not only to meet food safety requirements, but also as a policy response to promote healthy foods. These guidelines are intended to influence consumer behaviour , promote healthier food choices, and stimulate the production of healthier, nutritious foods by industry (Hawkes, 2004). In Ghana, food labelling is regulated by the Food and Drugs Authority (FDA), as established by law (Public Health Acts, 2012). The FDA guidelines for labelling of pre-packaged foods states that ‘all pre-packaged food products, whether locally- manufactured or imported to be offered as such to consumers, or for catering purposes must be labelled’. This guideline makes pre-packaged food labelling mandatory. However, the Nutrition fact panel (NFP), a component of food labels, is voluntary under Ghanaian law, except when a claim is made. The FDA labelling recommendations are based on and consistent with Codex Alimentarius Commission guidelines. University of Ghana http://ugspace.ug.edu.gh 4 While monitoring food label information on pre-packaged food products is a worthy public health goal, the plethora of pre-packaged foods available on the Ghanaian market and the multiplicity of the retail markets make this challenging. In this regard, the International Network for Food and Obesity/Non-Communicable Diseases Research, Monitoring and Action Support (INFORMAS) has developed a stepwise approach in surveying the food retail environment to monitor health-related label information(see figure 1) on pre-packaged foods (Swinburn et al., 2013). The INFORMAS approach harmonizes data collection across different countries to inform strategies for creating a healthy food environment and reduce obesity. 1.2 Statement of the Problem Consumer knowledge, attitudes, and perceptions of food and nutrition labels have received considerable attention as an important strategy for addressing poor diets and the obesity epidemic worldwide (WHO, 2006). As obesity and diet-related NCD prevalence rise in LMICs, consumer understanding and use of nutrition information on food labels can be useful for public health policy and action ( Hawkes et al., 2013). Majority of the research and recommendations in this area originate from developed countries. But in developing countries, limited attention has been given to research to understand consumer behaviour towards food label use in improving diets and preventing obesity (Mandle et al., 2015). To critically examine consumer’s use of food labels and how it translates into healthy dietary choices, examination of both consumer (demand-side) and product- related (supply-side) factors are important (Grunert, Fernández-celemín, et al., 2010; Hieke & Taylor, 2012). It is essential to analyse how nutrition information is communicated on labels of pre-packaged foods (Rayner et al., 2013). However, there is currently no empirical evidence on health-related label characteristics of pre- University of Ghana http://ugspace.ug.edu.gh 5 packaged foods, especially, the extent of penetration of food labels carrying nutrition and health claims in Ghana, as well as their utilization (Booth et al., 2021). Existing studies on consumer food label use in Ghana reports that consumers consult food labels (Ababio et al., 2012; Aryee et al., 2019; Azila-gbettor & Adigbo, 2013; Darkwa, 2014; Osei Mensah et al., 2012). However, not much is known about whether the consumer possesses the knowledge and skills that allow them to make healthy food choices as communicated on food labels. Evidence on consumers’ understanding of labels have been mixed (Booth et al., 2021). For instance, Aryee and his colleagues (2019) in their study assessing consumers food label use and understanding in Tamale, indicated that 66.7 % respondents claimed to understand food labels while Darkwa (2014) found only 22 % of consumers in Koforidua had adequate nutrition knowledge. These previous studies are limited since they employed subjective assessments in measuring consumer understanding and use of food labels. Besides, Ghanaian consumers are known to focus on expiry dates, however, use of health-related information such as nutrient composition of foods, ingredients list, health and nutrition claims are rare (Ababio et al., 2012; Osei Mensah et al., 2012). It is not known why the health-related aspects of food labels are not typically used by Ghanaian consumers. University of Ghana http://ugspace.ug.edu.gh 6 Figure 1: INFORMAS Taxonomy for describing health-related aspect of food labelling The bolded boxes represent ‘health-related food labelling’. Reference; Rayner, M., Wood, A., Lawrence, M., Mhurchu, C. N., Albert, J., Barquera, S., ... & L'Abbé, M. (2013). Monitoring the health‐related labelling of foods and non‐ alcoholic beverages in retail settings. obesity reviews, 14, 70-81. University of Ghana http://ugspace.ug.edu.gh 7 1.3 Justification In addressing the rising prevalence of NCDs, food labels constitute an important tool to help consumers make healthy food choices (Koen et al., 2016). Thus, this study is intended to fill the evidence gap on consumer understanding and use of nutrition information on food labels in Ghana. It also addresses a critical gap in research about the state of nutrition and health claims on pre-packaged foods in Ghana (Laar et al., 2020). Food labels constitute an important aspect of the food environment in its ability to aid healthy food choices. Despite its importance, very little is known about the type of nutrition and health-related food label information that consumers are exposed to. This knowledge is not existent in literature on Ghana, and addressing this knowledge gap is important against the backdrop of a rapidly changing food system (Andam et al., 2017; Andam et al., 2015). This study fills a significant gap in knowledge of Ghanaian consumers’ ability to use food labels to select healthy foods. Also, the study is important in extending our knowledge on the predictors of food label use beyond demographic characterizations such as the types of health-related information checked on pre-packaged foods, and to explore Ghanaian consumers’ reasons for non-use of food labels or otherwise. Finally, the evidence generated from this study will form part of the evidence needed to stimulate policies on food system reforms such as re-evaluation on nutrition labelling regulations, decisions on food marketing, healthier food product reformulations and consumer protection actions in Ghana University of Ghana http://ugspace.ug.edu.gh 8 1.4 Objectives of Study 1.4.1 General Objective The main objective is to assess label characteristics of pre-packaged foods and to determine the drivers of label use among consumers in urban Accra. 1.4.2 Specific objectives I. To describe and classify health-related nutrition information on labelled pre- packaged foods sold in local retail shops in Accra. II. To examine the level of compliance of health and nutrition claims information on labelled pre-packaged foods to Food and Drug Authority regulations and Codex standards. III. To determine pre-packaged food use behaviour among urban-dwelling households in Accra. IV. To characterize consumer perception and understanding of health-related information of labelled pre-packaged foods. V. To identify factors that are associated with consumer understanding and use of health-related label information on pre-packaged foods. University of Ghana http://ugspace.ug.edu.gh 9 CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 Food processing and unhealthy diets Advancements in food processing technology have increased the range of foods available, reduced cooking times, and improved the quality and safety of products. However, these benefits come with a cost on food systems, and may contribute to the increasing burden of chronic disease. A growing body of evidence has linked consumption of processed foods, especially ultra-processed foods (UPFs) with the global burden of non-communicable disease (Baker et al., 2020; Bielemann et al., 2015; Elizabeth et al., 2020; Fiolet et al., 2018). Although the relationship between food processing and health outcomes is still emerging, Hall and colleagues (2019) have suggested that consumers of ultra-processed foods showed increased total calorie intake and weight gain compared to non- consumers (Hall et al., 2019). In studies assessing dietary quality, a large share of ultra- processed foods, as part of the diet, are indicative of reduced dietary quality since they replace traditional whole and minimally-processed foods of populations (Luiten et al., 2016; Martínez et al., 2017). Industrial processing alters the chemical and structural properties of foods, thereby affecting the body’s metabolic response to food. Ultra- processed foods have been shown to elicit higher glycemic response and lower satiety than minimally-processed foods(Fardet, 2016). Further, processed foods containing certain food additives disrupt the normal gut microflora, increase gut permeability and induce autoimmune diseases(Lerner & Matthias, 2015). Observational studies have shown strong associations between increasing production and consumption of processed foods and increasing rate of obesity and other diet-related NCDs (Bielemann et al., 2015; Canella et al., 2014; da Costa Louzada et University of Ghana http://ugspace.ug.edu.gh 10 al., 2015). Also, an ecological assessment by Monteiro et al, (2018) in 19 European countries, showed that the household availability of ultra-processed foods was directly linked to the increasing prevalence of obesity(Monteiro et al., 2018). In prospective cohort studies, intake of ultra-processed foods were associated with the onset of cardiovascular disease linked with trans-fat, obesity, type 2 diabetes, hypertension and some cancers(De Souza et al., 2015; Fiolet et al., 2018; Mendonça et al., 2016, 2017; Rauber et al., 2015). Moreover, concerns about the health risks associated with consumption of processed foods and NCDs, have led to the development of food classification systems to differentiate between various categories of processed foods(Moubarac et al., 2014). In the extant literature, seven systems of classifying processed foods are mentioned and notable among them is the Nova food classification system (Sadler et al., 2021). The various criteria used in classifying processed foods include the extent of change of food from its natural state, the type of the change, where and who is making the change, the method whether traditional or modern methods and the reason or value of processing. The Nova system of food processing classification was developed by a group of researchers from the school of Public Health of the University of São Paulo.(Hall et al., 2019; Monteiro, 2009). Nova is based on the purpose, nature, and extent of processing of food or drink. The nova system classifies foods into four categories; minimally processed foods (MP), processed culinary ingredients (PCI), processed foods (PF) or ultra-processed foods and drinks (UPFDs)(Monteiro et al., 2010). UPFD foods rely on heavy industrial processing/formulations derived from whole foods. Examples include sugar, fatty or salty-rich packaged products, burgers, frozen pizza and pasta dishes, nuggets and sticks, crisps, biscuits, confectionery, cereal bars, University of Ghana http://ugspace.ug.edu.gh 11 carbonated and other sugary drinks. They usually contain less fiber or whole foods, high in salt, sugar, or fat. In recent years, there has been an increasing interest in the Nova system(FAO, 2015;Monteiro, et al., 2018). The Nova system has been widely used in studies assessing food security, diet quality and health outcomes such as obesity and diet- related NCDs (da Costa Louzada et al., 2015; Marrón-Ponce et al., 2019; Martínez Steele et al., 2017). Subsequently, it has influenced dietary guidelines of countries such as Brazil, Ecuador, Peru, Uruguay, Belgium and France(Herforth et al., 2019). The Nova framework has also received recognition from international organizations such as Pan American Health Organization (PAHO) and the FAO. However, the Nova system has come under serious criticism from the food industry and a section of the academic community(Gibney et al., 2017), challenging the evidence linking UPFDs with high dietary energy intake. Gibney et al, (2017), claim the evidence linking micronutrient malnutrition and UPFD consumption is weak. There is also criticism about the lack of considerations for portion sizes and rate of energy intake in the Nova classification as such not credible to inform the development of dietary guidelines (Gibney, 2019). These assertions had, however, been refuted with rebuttal publication from Monteiro et al, 2018, stating Gibney et al, (2017), refused to critically appraise the extant literature or probably doing the bidding for the Big Foods( Transnational food companies) (Monteiro et al., 2018). At the moment , there is no international agreement on what constitutes the levels of food processing(Sadler et al., 2021). Despite the growing evidence of the negative link between consumption of processed foods and health, little attention is being paid to food processing in public health, nutrition and epidemiological research especially in LMICs(Monteiro, et al., University of Ghana http://ugspace.ug.edu.gh 12 2018). Most processed foods and drinks offered come pre-packaged. Pre-packaged food products are vectors of these unhealthy diets. Worldwide, 75 % of the world's food purchases are processed pre-packaged foods and non-alcoholic beverages (Moodie et al., 2013;Popkin et al., 2011). Meanwhile, dietary habits have changed leading to increased consumption of pre-packaged foods(Baker et al., 2020;Popkin et al., 2013). Sales and rate of consumption of processed foods are increasing more rapidly in LMICs than in high- income countries (HICs)(Moodie et al., 2013). It is projected that the sales of ultra- processed foods in East and South Asia will be equal to those of high-income countries by 2035 (Haddad et al., 2016).These patterns are being driven by increasing globalization, technological advancement, industrialization of food systems, infiltration and activities of transnational food companies, and increasing economic growth (Moodie et al., 2013;Popkin, 2015; Stuckler et al., 2012). In Ghana, the food system is changing rapidly, with a notable shift in dietary behaviour towards increased consumption of processed pre-packaged foods, due to rapid urbanization and economic prosperity (Andam & Silver, 2016). Although the domestic production of processed foods is increasing, imported processed food dominate our market and their retailing is commonplace (Andam et al, 2016). There is also a boom in the supermarket industry associated with higher consumption of processed foods, and adversely, increases in the rates of obesity (Rischke et al, 2015). The supermarkets, public markets, shops and street hawkers offer consumers easy access to these processed foods. In Accra, it is easy for one to have home food shopping even in a car while driving home from work since street hawkers inundate consumers with a variety of packaged processed foods(Ofosu-Boateng, 2020). Consequently, there University of Ghana http://ugspace.ug.edu.gh 13 has been an increase in the availability, accessibility and affordability of pre-packaged foods(Dowuona-Hammond, 2018). 2.2 The burden of diet-related NCDs Globally, the prevalence of obesity and diet-related non-communicable disease (NCDs) continues to increase, especially in low- and -middle-income countries. NCDs account for the premature deaths of 15 million people annually and 85% of these deaths occur in low-and middle-income nations. (WHO, 2019). With already weak health systems in many parts of Africa and a high prevalence of communicable diseases, increasing numbers of patients with NCDs are putting unbearable pressure on health systems(Nyaaba et al., 2020). NCDs increase health expenses, thus worsening the plight of the poor and reducing nations’ productivity and economic growth (Bollyky et al., 2017; Chaker et al., 2015). Although the aetiology of NCDs is complex, four main risk factors of tobacco use, harmful use of alcohol, unhealthy diets and physical inactivity are responsible for the increasing prevalence of these chronic diseases (Michele Cecchini et al., 2015; Knai et al., 2018). However, dietary risk factors contribute more to the global burden of NCDs than tobacco, alcohol and physical inactivity combined (Danaei et al., 2014; Hyseni et al., 2017; Meier et al., 2019). Dietary risk factors(Poor diets) refer to high intake of calorie-dense foods, especially ultra-processed foods containing high amounts of sugar, salt, saturated and trans fat with the lesser intake of healthy foods like whole grains, pulses, nuts, fruit and vegetables (Melaku et al., 2016). Unhealthy diets are associated with increased risk of raised blood pressure, blood sugar, and lipids, as well as obesity in individuals. However, obesity represents a significant risk factor for the development of diabetes, cardiovascular diseases (CVD), osteoarthritis, sleep apnea, cancer, liver and kidney diseases(Pi-Sunyer, 2009). Over-weight and obesity represent University of Ghana http://ugspace.ug.edu.gh 14 excess calorie intake over the body’s energy requirements. As at 2013, little over two billion people were overweight or obese and 62 % of these individuals live in LMICs (Ng et al., 2014). In 2016, obesity was identified to be the world’s sixth-leading cause of disability-adjusted life years (DALYs) and the burden of obesity has been rising in LMICs (Ford et al., 2017). More recently, from 2010 to 2019 high body mass index ( BMI) was the leading risk factor of disability-adjusted life years(DALYs)(Murray et al., 2020). One feature of the obesity epidemic is, no country has been successful in reversing the increasing trend once it has begun. Obesity and diet-related NCDs have become a public health concern in Ghana (de Graft Aikins et al., 2012; Ofori-Asenso et al., 2016). Trend analysis of obesity data of Ghana Demographic Health Survey (GDHS) from 1993 to 2014, shows that the prevalence of obesity has increased from 3.4% to 15.3%(Ghana Statistical Service (GSS), Ghana Health Service (GHS), 2015). It is estimated that 43% of adults in Ghana are either obese or overweight (Ofori-Asenso et al., 2016). Despite the increasing prevalence of obesity and diet-related NCDs, there are no clear-cut comprehensive food policy approaches in Ghana aimed at improving healthy food choice and preventing diet-related NCDs (Allen et al., 2018; Laar et al., 2020; Nyaaba et al., 2020). 2.3 Food Labelling Policies addressing unhealthy diets and the obesity epidemic Fundamentally, food labelling policies were intended to inform and protect consumers about food products as well as ensure fair marketing. However, it has transitioned into a health policy tool to motivate change in consumer dietary behaviour and stimulate the production of healthy foods (Hawkes et al., 2013). On this basis, labelling that misleads and deceives consumers is frown upon by national legislations, international laws, and even private standards. Food label information should be truthful so that consumers can make informed decisions during purchase. Also, University of Ghana http://ugspace.ug.edu.gh 15 businesses are protected from unfair competition by the prohibition of false claims being made on products. Hawkes and her colleagues (2015), posited that food policies aimed at reducing the obesity epidemic should provide an enabling environment for healthy food choices, overcome barriers of healthy food choices, encourage consumers to select healthy food (empower the consumer) and provide the needed stimulus for the production of healthier foods (Hawkes et al., 2015). Similarly, the NOURISHING framework developed by WCRF (World Cancer Research Fund) identified the food environment, the food system and behavioural change communications as the main domains to comprehensively tackle obesity and diet-related NCDs (Hawkes et al., 2013). Since the causes of obesity are multifaceted preventive approaches must cut across individual, societal, environmental and economic aspects. Equally, a wide range of policies and regulatory interventions have been implemented over the years to prevent unhealthy diets and obesity in many parts of the world. Some notable actions include restrictions on unhealthy food advertisements especially to children, tax increases on sugar- sweetened beverages and mass media education to promote healthy eating and increased physical activity (Allen et al., 2018). Therefore, food labelling is well-fitted into this policy framework in addressing unhealthy diets and obesity. The provision of easy-to-understand nutrition information on pre-packaged foods would inform and empower consumers’ selection of healthy foods and incentivize the food industry to formulate healthier food products (Corvalán et al., 2013). Also, food labelling represents an important avenue for the ‘making the healthy choice the easier choice’ maxim as emphasized by many consumer watch groups in promoting healthy food choice (Ashe et al., 2011; Castres, 2016). University of Ghana http://ugspace.ug.edu.gh 16 Increasingly, food labelling has become one of the important policy areas in tackling unhealthy diets and obesity-associated NCDs in many parts of the world( Hawkes et al., 2013; Kanter et al., 2018; Stefan Storcksdieck genannt Bonsmann & Wills, 2012; Waterlander et al., 2017). Hence, WHO has recommended food labelling as one of its global policy packages to mitigate the rising burdens of diet-related NCDs (WHO, 2006). Likewise, the National Academy of Medicine (NAM) of the United States, the Organization for Economic Co-operation and Development (OECD) and the International Network for Food and Obesity/NCDs Research, Monitoring and Action Support (INFORMAS) have all identified food and nutrition labelling as an important policy area in addressing unhealthy diets and the obesity epidemic (Michele Cecchini et al., 2015; C Hawkes et al., 2013; McGuire, 2012; Swinburn et al., 2013). Furthermore, Bellagio conference (conference on programme me and policy options for preventing obesity in LMICs countries) reiterated the importance of nutrition labelling practices in addressing the obesity epidemic in LMICs (Popkin et al., 2013). There are, however, dissenting views about how food and nutrition labelling could help change the dietary behaviour of the population. Firstly, the knowledge on how labelling information aids a consumer to choose healthy foods is limited in many observational studies (Cowburn & Stockley, 2005; Drichoutis et al., 2006; Grunert & Wills, 2007). Although eye-tracking technology was employed to bring the needed objective measurement in food labelling research to mimic real-world shopping, it is limited by the use of unrepresentative samples for such study designs (Graham et al., 2012). Barker et al, (2012), question the importance of labels and argued that simply providing easy-to-understand information can not directly translate into much improvements in diets (Barker et al., 2012). Similarly, review of literature assessing University of Ghana http://ugspace.ug.edu.gh 17 effectiveness of various nudges towards healthy eating , label information-cognitive nudge- was found to have the least impact compared with affective and behavioural nudges(Cadario & Chandon, 2019). These studies demonstrated that there is a disconnect between what consumers with nutrition label knowledge do in real shopping experience and what they claim they do. Mhurchu and Gorton, 2007 in their review of nutrition labelling and claims in New Zealand, contended that the provision of nutrition labels in an easy-to-understand and user-friendly manner provides a supportive environment that empowers people to make healthy food choices (Mhurchu & Gorton, 2007). Magnusson (2010) argues that food labelling is only a part of a whole range of policies needed to make gains in obesity prevention and public health which has a potential to empower the consumer and act as catalyst for production of healthier foods(Magnusson, 2010).Also, in analysing the effect of different types of labels on nutritional quality of supermarket food purchases, Dubois et al.(2021) point outs that the marginal effect on nutritional quality must be situated in the right perspective since small changes in the nutritional quality of diets over a long period can have significant health outcomes (Dubois et al., 2021). Moreover, it is argued that a strong legislative and regulatory environment coupled with nutrition education has the potential of changing the dietary behaviour of populations(Michele Cecchini et al., 2010; Mandle et al., 2017). The provision of easy- to-understand label information coupled with education on nutrition would lead to improvement in individual dietary behaviour. Implementing an effective food labelling policy is in tandem with the axiom “making the healthy choice an easy choice’’ for consumers purchasing pre-packaged foods. In this way, food labelling will have a greater impact on addressing unhealthy diets and the growing epidemic of obesity and diet-related NCDs (Volpp & Asch, 2017). Despite these benefits of food labels to the University of Ghana http://ugspace.ug.edu.gh 18 individual and public health as a whole, the prevalence of actual use of food labels has been lower than self-reported use especially in LMICs (Cowburn & Stockley, 2005). 2.4 Food and nutrition labelling of Pre-packaged Foods and their Regulation 2.4.1 Food Labelling of pre-packaged foods a standard of Codex Information on food labels is generally set and standardized by Codex. The Codex Alimentarius Commission (Codex) was established by the Food and Agriculture Organization and the World Health Organization in 1962 with a core mandate for developing food standards. They also provide guidelines for label information and food groups. The aspects include standards on general labelling of pre-packaged foods (Codex Stan-1985), labelling of food additives ( Codex Stan 107-1981), labelling of and claims for pre-packaged foods for special dietary uses and foods of medical purposes(Codex Alimentarius Commission, 2001). Also, they provide guidelines on nutrition labelling and claims as well as their uses. Additionally, they provide guidelines on labelling information of production, processing, marketing and religious indications on pre-packaged foods (CAC/GL 23-1997, CAC/GL 24-1997, CAC/GL 32- 1999). Generally, Codex guidelines are fundamentally based on the principle that any information provided on labels should not be false, misleading or deceptive for the consumer. According to Codex, food labelling is defined as ‘any written, printed or graphic matter that is present on the label, accompanies the food, or is displayed near the food, including that to promote its sale or disposal’.(Codex Alimentarius Commission, 2001). Likewise, Codex defines a food label of a pre-packaged food as ‘ any tag, brand, mark, pictorial or other descriptive matter that is written, printed, stencilled, marked, embossed or impressed on, or attached to, a container of food product’(FAO, 2016). Pre-packaged food and drink are defined as ‘any food item for University of Ghana http://ugspace.ug.edu.gh 19 presentation to the consumer or caterer made in advance in a container including wrappers such that packaging completely or partially encloses the food item such a way the contents cannot be altered without opening or changing the packaging’ (Codex, 2012; FDA, 2013b). A wide range of pre-packaged foods carries labels from minimally processed to ultra-processed food products. Food labels contain a lot of information but the food identity (name of food product, lot number, address of the manufacturer, country of origin, its net weight), ingredients list, food additives, allergenic ingredient declarations and date markings are minimum information requirements that must be provided on labels. These requirements are mandatory for food labelling in most countries. Information on nutrient composition and provisions of claims are voluntary indications on pre-packaged foods in many countries(FAO, 2016). Nevertheless, nutrition labels, ingredients list and claims are aspects that convey nutrition and health information intended to guide selection of food products. These components of labels are often the subject matter of food label research and government regulations (Miller & Cassady, 2015; Rayner et al., 2013). Food labelling practices are regulated by many nations based on codex standards and guidelines. 2.4.2 Nutrition and Health Information on Pre-packaged Foods Nutrition labelling is the disclosure of information on food composition, nutrient constituents, and quantities in food products. Therefore, ingredient list, nutrition fact table and supplementary nutrition information (use of health and nutrition claims) constitute nutrition and health information on food labels. Essentially, the provision of these aspects of the label characterizes how consumers are informed about the contents of foods, the benefits and risks associated with nutrients of public health significance. In this way, a consumer is empowered to make an informed food choice. University of Ghana http://ugspace.ug.edu.gh 20 Currently, nutrition display formats on pre-packaged foods fall into two general categories: Back- of- Pack (BOP) and Front-of-Pack (FOP). These aspects of nutrition labelling are discussed below. 2.4.2.1 Nutrition Fact Table (Back-of-Pack) A typical Nutrition Fact Table (NFT) consists of a nutrient list, their relative amounts and other forms of numerical quantifications such as serving size and percent daily values of macronutrients, vitamins and minerals (see figure 2). It is usually displayed at the back of package, and it is often referred to as Back-of-Pack (BOP) or Nutrition Fact Panel (NFP). The NFT is the most commonly used nutrition label format worldwide (Temple et al., 2014) however there are variations in the way information is presented across nations due to different regulatory controls. In the US, the enactment of Nutrition Labelling and Education Act (NLEA), some three decades ago made nutrition labelling mandatory on pre-packaged foods. Currently, codex recommends mandatory nutrition information of proteins, available carbohydrates, saturated fat, total fat, sodium and sugars total calories, serving sizes and nutrients relative percent daily values (Codex, 2012). Percent daily values are based on nutrient reference values (NRV). NRVs are a set of numerical values established for nutrition labelling based on scientific data associated with nutrient requirements or with reducing the risk of diet related-NCDs(Codex, 2012). Auditing nutrition labelling on pre-packaged foods in countries with high consumer awareness of labels and mandatory nutrition labelling, showed a high rate of nutrition label presence. For example in the US, over 98% of NFT were on pre- packaged foods on the market, averagely 85% of products have NFT in Europe, 88% in China and 96% in Australia(Bonsmann et al., 2010; Huang et al., 2016; Legault et al., 2004; Sussman et al., 2019). However, in the analysis of pre-packaged nutrition University of Ghana http://ugspace.ug.edu.gh 21 labelling practices in some LMICs, in Malawi, 40.4% of products sampled had nutrition declaration, 65.9% in Serbia and 70% in Slovenia (Davidović et al., 2015; Kasapila & Shaarani, 2013). The nutrition fact table is an important information source to help consumers compare the nutrient profiles of different processed foods, ascertain claims and select products to meet their dietary requirements. However, consumer use of this component has several limitations. The issue of low visibility(found at the back of pack), the complexity of information especially quantitative information makes it difficult for consumers to comprehend and some consumers find its details too overwhelming (Campos et al., 2011; Grunert, Fernández-celemín, et al., 2010). Thus, across the world, NFTs have been shown to be inadequate in helping consumers, especially those with low nutrition literacy in selecting healthy foods (Anastasiou et al., 2019; Campos et al., 2011; Cowburn & Stockley, 2005). University of Ghana http://ugspace.ug.edu.gh 22 Figure 2: Examples of Nutrition label BOP adapted from(Campos et al., 2011) University of Ghana http://ugspace.ug.edu.gh 23 2.4.2.2 Ingredient List The ingredient list component conveys information on constituents used to make food products. For some food packages, it is part of the nutrition fact table. Except for single-ingredient item foods, provision of the ingredient list on pre-packaged foods are mandatory. This aspect of labels helps consumers to assess the nutritional and health value of foods. Consumers can identify processed foods with significant amounts of healthy ingredients. For instance, it can be used to assess the level of synthetic additives (e.g. tartrazine- synthetic colour- is in sweetened beverages) and ingredients that must be avoided or reduced in food products such as saturated fat, added sugar or salt. In a focus group study of label use in South Africa, participants cited allergic reactions to certain food additives drove them to pay attention to ingredient lists on pre-package foods (Kempen et al., 2011). Regulatory authorities recommend various specifications regarding the content, the style and the formats of how ingredients should be listed. These recommendations are meant to aid consumers in their evaluation of the food items (FAO, 2016). For example, ingredients are listed in descending order of their quantities by weight so smaller ingredients are towards the end of the list. Even though font sizes are specified for easy readability, most studies indicate font sizes are barriers to the use of ingredients (Campos et al., 2011; Mackey & Metz, 2009). Nutrition information provided on labels must be consistent with national dietary guidelines. For example, the 2020-2025 US Dietary Guidelines states that; “ increase whole grains in place of refined grains by using the ingredient list on packaged foods to select foods that have whole grains listed as the first grain ingredient”(USDA, 2020). Despite most studies include ingredient list as an information source, some studies ignore their usage in assessing consumer nutrition label use (Miller & Cassady, 2015). University of Ghana http://ugspace.ug.edu.gh 24 2.4.2.3 Front-of-Pack Labels (FOP) FOP nutrition declaration system is an effort to enhance the BOP format for easy consumer understanding and use. Here, information is presented on the front of food packages in simple and easy-to-understand formats allowing consumers to make a quick decision about the nutritional content and relative healthiness of food products (see examples in Figure 3). FOP nutrition labelling is also seen as a nudge policy approach in creating a healthy food environment (Scrinis & Parker, 2016). The formats are designed in a way to lead consumers towards healthy foods as much as possible. There are two main identifiable groups of FOP systems in literature (Dean et al., 2014; Roseman et al., 2018). They include nutrient-specific and summary systems. The nutrient-specific system provides nutritional information of several nutrients as a guide such as Guideline Daily Amounts (GDA) in the USA, the Multiple Traffic Lights (MTL) in the UK (Scrinis & Parker, 2016) and warning labels in Chile. More recently, Mexico has also adopted the FOP warning label system replacing the GDA(White & Barquera, 2020). The summary system tends to display information about the whole nutritional quality of the food product based on a nutrition profiling system, examples include the keyhole symbol in Sweden, the choice logo in the Netherlands, guiding health stars in Australia and 5-colour nutritional labels in France (Julia et al., 2017; Khandpur et al., 2018). Most of the FOP systems developed originated from food industry associations and non-governmental organizations. However, development of some FOP labelling systems was the collaboration between industry efforts, the research community, consumer groups, governments and non-governmental organizations. Initially, FOPs were implemented voluntarily by food industry associations. However, in quest of promoting healthy diets, governments have led the implementation of FOP systems in University of Ghana http://ugspace.ug.edu.gh 25 many countries. For example, in the UK, the Traffic Light Labelling system idea was mooted by the Non-Government Organisations(NGO) Coronary Prevention group and the choice programme was an international industry-led action before governments took over (Dean et al., 2014). For the past decade, the multiple traffic light (MTL), the daily amount (GDA) and choices logo FOPs have dominated the European market space (Pauline et al., 2015). In a study on GDA preference, Grunert et al, (2010) reported that respondents prefer GDA in their assessment of food products' healthiness. In the United States, middle-to high-income consumers reported that multiple traffic light systems were their preferred format in identifying and evaluating nutritional profiles of foods(Gorski Findling et al., 2018). In a more recent study, warning label was widely reported as the best understood and preferred FOPs than other FOPs and NFT(Nieto et al., 2019). However, in systematic review and meta-analysis of randomized studies, MTLs were found to be more effective in helping consumers choose healthier foods than other labelling formats(M. Cecchini & Warin, 2016). More recently, the FOP formats, Multiple traffic lights and warning systems have been shown to lead consumers to healthy products (Emrich et al., 2017; Neal et al., 2017). Up to date, there is no international agreement as to which FOP labelling system is most helpful in guiding consumers to choose healthier foods. Several studies have reported the potential benefits of FOP labelling systems influencing population diets(M. Cecchini & Warin, 2016; Pettigrew et al., 2017; Rønnow, 2020; Scrinis & Parker, 2016). The evidence of FOP labelling impact is stronger in its influence of the supply side; that manufacturers’ reformulation of healthier foods than the demand side-consumer behaviour (Vyth et al., 2010). Thus, there is a gradual global trend towards FOP nutrition labelling with many countries adopting various formats-fig. 3- (Kanter et al., 2018). Currently, there are no clear-cut University of Ghana http://ugspace.ug.edu.gh 26 indications and guidelines for national mandatory FOP nutrition labelling (Kanter et al., 2018). Moreover, the FOP system has been criticized by others as a mere marketing strategy of food industries to increase profits rather than a public health policy tool(Brownell, 2012; Brownell & Koplan, 2011). In other situations, the implementation of the FOP system and its regulations have led to legal tussles between the food industry and governments (Jones et al., 2019). For example, due to the implementation of warning labels in Chile, Pepsico sued the Chilean Treasury for the prohibition of certain features on their food packages as it infringed on its intellectual property rights (Aguayo, 2017). Also, the European Commission initiated infringements proceedings against the UK regarding the ‘traffic light’ labelling scheme (Salas & Simões, 2014). University of Ghana http://ugspace.ug.edu.gh 27 Figure 3: Examples of FOP labelling systems adapted from Chantal, J., Hercberg, S., & World Health Organization. (2017). Development of a new front-of-pack nutrition label in France: the five-colour Nutri-Score. Public Health Panorama, 3(04), 712-725. University of Ghana http://ugspace.ug.edu.gh 28 2.4.2.4 Health and Nutrition Claims Another aspect of nutrition labelling that provides the consumer easy-to- understand information is the use of claims especially claims relating to nutrition and health. Health and nutrition claims are representations either presented in words, numbers or pictures that relate to unique nutrition or health properties of the food product. Their definitions and their classifications vary from one jurisdiction to another. However, Codex guidelines identify ‘nutrient content claim’ and ‘nutrient comparative claim’ as two groups of nutrition claims. Also, ‘nutrient function claim’, ‘reduction of disease risk claim’ and ‘other functions claims’ are the types of health claims in the standards of the Codex. Codex guidelines recommend that health and nutrition claims should be consistent with national nutrition policy. Besides, claims allowed should be substantiated by sound scientific evidence. INFORMAS provided elaborate and more definitive definitions of claims in general (see figure 1). Their definitions are based on the different types of claims proposed by Codex (Rayner et al., 2013). Among the sub-categories of health and nutrition claims, nutrient content claims are most common on labelled food products. They are provided on a voluntary basis. They usually relate to calories, protein, carbohydrate, fat, sodium, vitamins and minerals which NRVs values have been established in the standards of codex. Examples of nutrient content claims’ phrases include low in, high in, source of, no added, source of and fortified/enriched with. Although the provisions of claims on package foods were commonplace in wealthy nations, their presence has become a feature of retail markets in developing countries. Arguably, it has become a marketing tool for food manufacturers to boost their sales. Research has shown that health and nutrition claims on packaged foods can influence consumer evaluation of the healthiness of the products and can lead to an University of Ghana http://ugspace.ug.edu.gh 29 increase in the sales of products (Anastasiou et al., 2019; A. Kaur et al., 2017). On the other hand, they may lead consumers to overrate the healthiness of products and could lead to overconsumption. The use of claims could also mislead consumers in their food choice by showcasing beneficial aspects of the product while hiding lesser desirable attributes. In monitoring the US food supply, it was identified that 43.1% of new pre- packaged products introduced on the market had one form of HNCs claims or the other (Martinez, 2013). In assessing the prevalence of HNC in one of the largest grocery supermarkets in the UK, 32% of the pre-packaged food products either had health and nutrition claims (A. Kaur et al., 2016) and an average of 26% was identified from food products across five European countries (Hieke et al., 2016). In a similar study in Mongolia, only 9% of all food product samples had at least a nutrition or health claim (Chimedtseren et al., 2020). In Africa, a study to analyse the presence of claims on pre- packaged foods supply, 14% of HNC was identified on pre-packaged foods produced in Malawi, 20% for South Africa and 36.6% from other southern African countries (Kasapila & Shaarani, 2013). Inherent in food laws across many nations; food products presented to consumers should not be false, misleading or deceptive. In general, provisions of nutrition and health claims on packaged are meant to be substantiated in the nutrition fact table of label information or are to be part of the approved list from regulatory bodies. Therefore, governments have the responsibility to monitor, regulate and enforce laws to ensure the credibility of claims made on labelled pre-packaged foods. 2.4.3 Voluntary and Mandatory Nutrition Labelling of Pre-packaged foods Globally, regulatory controls for nutrition labelling for pre-packaged foods had mainly focused on nutrition fact tables and their contents (BOP nutrition labelling). The University of Ghana http://ugspace.ug.edu.gh 30 legislative and regulatory controls for nutrition labelling either fall into mandatory or voluntary controls. In the review of Hawkes et al, (2004), countries can be grouped into the following categories of nutrition regulatory policies. - Mandatory requirements of Nutrition Fact Table/NIP on some or all pre- packaged foods. - Voluntary requirements of NFT unless a health or nutrition claim is made. - Voluntary requirement excerpt in the case of food with special dietary uses like baby foods, diabetic foods, fortified enriched foods. - Voluntary requirements but prescribes standards in case of use. Essentially, nutrition labelling regulation can be viewed as either mandatory or voluntary requirements, with variations in regulatory provisions for the format and content indications of nutrition labels across different countries. Disclosure of nutrition information on pre-packaged foods is voluntary in most nations. Under the voluntary regulation regime, producers are permitted to present nutrition information in any format acceptable. There are no restrictions to which type of product that should have nutrition information. However, in some jurisdictions, declaring a health or nutrition claim requires the provision of NFT. For example, Botswana, Malawi, Mozambique, Namibia, Tanzania, Zambia, Zimbabwe, Egypt, Kenya and including Ghana subscribe to voluntary nutrition labelling (EUFIC, 2018). Although voluntary labels allow food producers to be innovative about communicating nutrition information, it does not allow consistency in the marketplace to enable consumers to compare nutrition profiles of different products. Also, under voluntary schemes, producers can hide undesirable attributes of their products. Audit analysis of pre-packaged in countries with voluntary nutrition labelling , showed low standard of labelling practices were observed including; low compliance to labelling University of Ghana http://ugspace.ug.edu.gh 31 standards and insufficient nutrition labelling (Chimedtseren et al., 2020; Kasapila & Shaarani, 2013; Wang et al., 2011). In Ghana, presently, it is unclear whether nutrition declarations, as well as provisions of health and nutrition claims on food labels, are monitored and evaluated to understand the kind of nutrition information consumers are exposed to (Laar et al., 2020). Mandatory labelling policies had been a key regulatory feature of developed countries like the USA, Canada, Australia, Japan, the European Union but in recent times, a lot more LMICs are turning to mandatory nutrition labelling (EUFIC, 2018). For example, Nigeria recently turned to mandatory nutrition labelling for all pre- packaged foods. Mandatory nutrition information disclosure allows for a standard format that enables consumers to easily compare nutrient profiles of food products. Also, under mandatory labelling more food products disclose nutrition information than voluntary labelling. It requires full information disclosure about food products enabling producers to even disclose otherwise undesirable information about their product. This mandatory requirement can stimulate producers to reformulate healthier food products (Hawkes et al., 2015). With interest in FOP labelling, which is mainly regulated voluntarily, it is possible to identify both voluntary and mandatory information on single packaged food. Regardless of whether a country adopts either voluntary or mandatory nutrition labelling, efforts must be made to continue to align labelling policy towards informing and empowering consumers towards healthy food choice (Jones et al., 2019). University of Ghana http://ugspace.ug.edu.gh 32 Figure 4: Timeline of countries all over the world adopting FOP labeling system. Adopted from (Kanter et al., 2018) University of Ghana http://ugspace.ug.edu.gh 33 2.5 Consumer food label use Food label use: a behaviour of how consumers engage with food label information is defined in a variety of ways across studies. Four main concepts were identified in literature. - Self-reported food label use -Food label understanding through self-reports or objective test that measures actual understanding -The tracking of retail data to see how food label use affects purchase -Longitudinal and modelling approaches in measuring changes in dietary consumption patterns with label use (Campos et al., 2011; Mandle et al., 2015) However, self-reported food label use studies dominate the literature (Campos et al., 2011). Hence, self-reported studies indication of label use is typically higher above 50% than studies that employed objective assessments of respondents(Campos et al., 2011; Cowburn & Stockley, 2005). Lower rates of understanding of labels are reported among respondents who were assessed objectively (Grunert, Fernández- celemín, et al., 2010). Besides, operationalizations of the term ‘food label use’ also is defined differently across studies making comparisons difficult. In most cases, the search for and the reading of nutrition labels within certain time frames are identified as ‘label users’. In analysing cross-sectional data on the use of labels, Christoph and colleagues, 2018 identified a ‘label user’ as respondents who responded ‘ most of the time’ and ‘always’ to the question ‘How often do you use food label: ingredient list, serving size information , nutrition and claims before buying or choosing to eat a food product for the first time?’(Christoph et al., 2018). Other responses are deemed ‘non label users. University of Ghana http://ugspace.ug.edu.gh 34 In the study of assessing consumer understanding and use , Jacobs et al, employed a food labelling task which is seen to be a more objective assessment among consumers(Jacobs et al., 2011). While such assessments are seen to be more objective than self-reports they are also limited because they do not mimic the real world of shopping and affected recall bias. Therefore, an approach that seems to answer the research question of how nutrition label use can translate in healthy dietary choices is the use of eye-tracking cameras. These precise eye-tracking cameras monitor consumer visual attention and measures how consumers view and evaluate labels during shopping. Food label use prevalence (in terms of frequency) was found to be 82% in New Zealand (Mhurchu & Gorton, 2007), 52 % in Canada, 47 % in EU (Grunert, Fernández- celemín, et al., 2010) and 75% in the USA (Campos et al., 2011) according to their national representative data. For other non-representative studies, it was 40.5% in Lesotho (Mahgoub et al., 2007), 48 % in South Africa (Bosman et al., 2014), 63.2 % in United Arab Emirates (Basarir & Sherif, 2012) and a study found 80.8 % use prevalence in Nigeria (Oghojafor et al., 2012). Despite the numerous studies on consumer label use, most studies rely on convenience samples and a section of the population limiting our understanding of the predictors of food label use (Soederberg & Cassady, 2015). The situation is not different from Ghana where five studies on labelling identified, is only one that attempted to use random samples (Darkwa, 2014) and the rest use non-representative samples (Ababio et al., 2012; Aryee et al., 2019; Azila-gbettor & Adigbo, 2013; Osei Mensah et al., 2012) . It is difficult to generalize such results to the larger population. Also, not much of the situation is known about Accra since other studies were done outside Accra (Darkwa, 2014). University of Ghana http://ugspace.ug.edu.gh 35 2.6 Reasons for use and non-use of nutrition information on food labels At the end of the international conference on Nutrition dubbed, Rome Declaration (2014), it was recommended that all countries should endeavour to empower consumers and create healthy food environments. Food labelling as a policy tool has the potential of a wider reach, a good source of health-related information of food products, less restrictive and low cost than strategies of using subsidies and taxes. Therefore, it is important to explore the reasons for non-use and use of nutrition information to inform labelling strategies to create the awareness, ensure effective use and empower the consumer toward healthy food choice. Historically, investigators had attempted to answer this research question in the US after food labels became widely available (Klopp & MacDonald, 1981). For instance, Klopp and MacDonald, (1981) in exploring reasons for consumer reasons for non-use identified three categories of reasons for low rate of use. The reasons offered were related to three categories of consumer behaviour: shopping practices, the absence of need and perceived inability. The study identified 79% of non-users indicating that they trusted in their ability to choose a healthy food without consulting food labels (absence of need), few others complained of time constraints when shopping and a smaller number explained that nutrition information is confusing (perceived inability). In a more recent study in South Africa , consumers who do not use labels ranked price and taste as more important than nutrition information (Jacobs et al., 2011) however for label users the motivation to use health stem from reasons relating to the product and their health. Besides, most label users said they read food labels to know the level of nutrient content in food and product quality. In a cross-sectional survey of consumers in Madrid Spain, the investigators identified time constraints (38.9%), absence of need University of Ghana http://ugspace.ug.edu.gh 36 (27.1%), and reading problems (18.1%) as common reasons given for not using information on food labels (Prieto-Castillo et al., 2015). In Rothschild’s(1999) conceptual framework, observed differences in behaviour of consumers were explained by the levels of motivation, opportunity and ability of individual consumers (Rothschild, 1999). The influences of and the interplay of sociodemographic factors including literacy and numeracy skills, the type of food environment, knowledge of nutrition issues explained whether the consumer has the motivation to search for, read, understand and use label information to guide his or her purchase. The rest of the sections discussed the influences of these factors. 2.7 Consumer perception and understanding of food labels Food choice can be a complex decision to make. The food choice process model states that an individual decision to select food is based on the negotiations of taste, price, health, convenience and social reasons (Furst et al., 1996). Label information is an important tool to raise awareness of nutrition and health implications linked with pre-packaged food consumption. Some research has revealed that most consumers consider nutrition labels as an important information tool however their understanding is seen to be poor (Campos et al., 2011; Cowburn & Stockley, 2005). Understanding food labels is key to the effective use of labels (Hieke & Taylor, 2012). According to behavioural models on the consumer decision-making process and attitude formation, perception is a strong precursor to understanding (Grunert & Wills, 2007; Jacobs et al., 2011). Individual perception is either conscious or subconscious with conscious perception exerting a stronger effect on subsequent behaviour. Therefore, how information is perceived determines the kind of meaning attached. The meaning can either be subjective or objective. Subjective understanding is the kind of meaning an individual perceives information is, but objective understanding is the kind University of Ghana http://ugspace.ug.edu.gh 37 of meaning an individual perceives about the actual truth. Motivation in, interest, and knowledge of nutrition issues are expected to be stronger determinants of how label information is perceived, understood and use (Campos et al., 2011; Drichoutis et al., 2005). As often pointed out, individual demographics are described as predictors of label use, but in actual terms, they are predictors of motivation, interest and nutritional knowledge (Grunert, Wills, et al., 2010). Therefore, motivation of, interest in, and knowledge of nutrition and health are the immediate predictors of food label use. Generally, observational studies suggest that the younger population, those with higher education, good literacy and numeracy, with better socioeconomic status and nutrition knowledge were more likely to report positive perception and understanding of nutrition labels (Koen et al., 2016; Mandle et al., 2015). With studies that employed a more objective assessment of understanding requiring respondents to perform a labelling task, the understanding was greater in younger adults, being female, educated, with adequate subjective understanding, high interest and appreciable knowledge in nutrition, positive attitude towards and motivation to use the nutrition labels (Grunert, Wills, et al., 2010; Liu et al., 2015; Mandle et al., 2015). 2.8 Factors associated with Consumer food label use In the critical review of nutrition labelling, Hieke and Taylor, (2011) opined that to investigate factors associated with consumer food label use, it is important not only to analyse demographic factors but examine label characteristics (the type of information and their manner of presentation) consumers are exposed to. Generally, in the extant literature, studies are either examining consumer characteristics with little or no involvement of the food product information consumers are exposed to or vice versa. University of Ghana http://ugspace.ug.edu.gh 38 Research on the examination of label characteristics can be divided into two broad categories; format and wording (Hieke & Taylor, 2012). The focus has largely been on nutrition information and claims. Research has shown that nutrition information of BOP labelling is not easily comprehensible for most consumers, especially those with little or no knowledge of nutrition as compared to FOP labelling(Mandle et al., 2015). Wansink (2003) using FOP and BOP labels demonstrated that short claims of FOP combined with more detailed provision at the back of the package led consumers to a better decision about healthy foods (Wansink, 2003). On the other hand, regarding label wording, most consumers prefer verbal or pictorial (qualitative) information than numeric (quantitative information) (Howlett & Kennedy, 2011). Regarding factors associated with food label use, the literature has focused on two main aspects; personal and socio-demographic factors (Campos et al., 2011). Personal factors refer to ability (efficacy), motivation and the knowledge levels of individuals to search, read and understand food labels at the point of purchase. According to Barreio-Hurle (2010), nutrition knowledge is the strongest predictor of food label use since consumers with higher knowledge are motivated and are capable of using the information to choose healthy foods (Barreiro-Hurl, Gracia, & de- Magistris, 2010). Socio-demographic factors such as age, gender, family size, education, income/ occupation, religion and ethnicity/race are shown to be associated with label use (Campos et al., 2011; Hieke & Taylor, 2012; Jamal & Sharifuddin, 2015; Mandle et al., 2015). However, in extant literature education and income levels seem to be stronger predictors of consumer food label use than other socio-demographic factors (Campos et al., 2011; Grunert, Fernández-celemín, et al., 2010).In a recent systematic review, University of Ghana http://ugspace.ug.edu.gh 39 Anastasious and colleagues (2019), found that food label use was associated with overall improved diet quality, reduced calorie intake and increased consumption of fruits and vegetables (Anastasiou et al., 2019). Ethnic and religious differences have been observed to be associated with food label use in terms of the type of information sought. For example, in a study among ethnically diverse shoppers in New Zealand, low-level food label usage was observed among minority ethnic groups (Gorton et al., 2009). Also, some consumers’ religious belief informs their dietary choices such vegetarians who watch out for no trace of meat/fish in a product, Muslims watch out for ‘halal’ symbols on meat products while Judaists watch out for ‘kosher’ (Jamal & Sharifuddin, 2015; Zepeda et al., 2013). 2.9 Conceptual Framework The conceptual framework for this study (Figure 5) is an adaptation of Grunert and Wills, (2007) and Jacobs et al., 2011 conceptual models (Grunert & Wills, 2007; Jacobs et al., 2011). The combination of the two frameworks help maps out, identify pathways and explain linkages of consumer understanding and use of food label information. The theoretical framework of Grunert and Wills (2007), posited that the understanding and use of labels are based on consumer choice behaviours and the drivers of choice. Jacob et al, (2011), distinguished these drivers as external/supply factors (including product attributes, the food label information, role of food manufacturers and food labelling regulation) and internal/demand factors (factors about individual demographics characteristics and their situational factors). The goal of this study was to assess label characteristics of pre-packaged foods and to determine the drivers of label use among consumers in urban Accra. Grunert and Wills (2007), model provided a pathway to analyse consumer decision making and food University of Ghana http://ugspace.ug.edu.gh 40 label understanding and use behaviour while Jacob et al., (2011) framework identify the drivers of label use, as such the current conceptual framework is apt for this study. In this conceptual framework, the outcome measure is consumer behaviour towards food labels on pre-packaged foods (the middle rectangle in Figure 5). However, consumer behaviour towards food label was operationalised to measure consumer food label use and understanding as the main dependent variables for this study. The internal factors (independent variables) are those that fall within the consumers’ power and influence their decisions on label use and understanding. However, external factors influence the type of information provided and communicated on food labels to the consumer (Balasubramanian & Cole, 2002). Both the external and internal factors are important determinants of consumer comprehension and use of labels, and consequently, consumers’ ability to make informed pre-packaged food choices. Consumer decision-making is based on the processes that influence food product selection in the face of multiple options and how label information available on each alternative affects their selection (Figure 5). Therefore, consumers’ motivation to buy certain food products might be due to the label information they are exposed to, and their perception of product’s information. The probability of exposure is increased if consumers search for label information to read. However, the exposure would lead to how the information on the product is perceived. How the product is perceived would inform consumer appreciation (liking) and their understanding of the label information on the product. Moreover, for label use to drive better decision-making, understanding the product’s label is important. This is particularly relevant, in respect of consumer processing of label information, and subsequently, its effects on decision-making regarding the product (Cowburn & Stockley, 2005). Hence, differences in consumers’ University of Ghana http://ugspace.ug.edu.gh 41 nutrition knowledge, demographic characteristics and situational factors influence understanding and food label use. Situational factors include time, ability to prioritize, nature of work, health status, and socioeconomic status. However, with appropriate nutrition knowledge, a consumer can be motivated to select healthy food products. In assessing consumer understanding, it is important to differentiate between subjective and objective understanding (Grunert, Wills, et al., 2010; Jacobs et al., 2011). Subjective understanding refers to the meaning the consumer ascribed to information and their perception of the information. Objective understanding, on the other hand, refers to the intended meaning of the information provided by the manufacturer, and the extent to which the consumer appreciates this intended message. It is important to note that liking a product may not necessarily mean understanding the message on the label. It could also mean that consumers appreciate other characteristics of the product, for example the symbols and colours of the label or packaging. This perception could lead to a positive evaluation of the food product, which will be misleading, if it is not linked to the understanding of nutrition information provided. Petty and Cacioppo, 2012, refer to this type of information evaluation as peripheral information processing(Petty & Cacioppo, 2012). University of Ghana http://ugspace.ug.edu.gh 42 Internal Influences • Demographic characteristics Gender Age Household size Education Health Status Nutrition Knowledge • Situational Status Work Status Income Time External Influences • Product Attributes Price Taste • Food Label Information Ingredient List Nutrition information Food claims Expiry/manufacture dates • Role of Food Manufacturers • Food-Labelling regulation Compliance of nutrition and health claims Need/Motivation Search Exposure Perception Food label Use Liking Understanding Informed Food Choice Consumer Benefits Recommendations Figure 5: An adapted Conceptual Framework of consumer understanding and use of food label University of Ghana http://ugspace.ug.edu.gh 43 2.10 Limitation of the Literature Review Although a critical literature review was undertaken, it is limited in the light of the fact that a