Intake of Omega-3 Long Chain Polyunsaturated Fatty Acid (N-3 LCPUFA)

University of Wollongong Research Online

University of Wollongong Thesis Collection University of Wollongong Thesis Collections

2013 Intake of omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) among Australian children (2-16 years): in depth nutrient and food analysis Setyaningrum Rahmawaty University of Wollongong

Recommended Citation Rahmawaty, Setyaningrum, Intake of omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) among Australian children (2-16 years): in depth nutrient and food analysis, Doctor of Philosophy thesis, School of Health Sciences, University of Wollongong, 2013. http://ro.uow.edu.au/theses/3945

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library: [email protected]

INTAKE OF OMEGA-3 LONG CHAIN POLYUNSATURATED FATTY ACID (n-3 LCPUFA) AMONG AUSTRALIAN CHILDREN (2-16 YEARS): In depth nutrient and food analysis

A thesis submitted in fulfilment of the requirements for the award of the degree

DOCTOR OF PHILOSOPHY

from

UNIVERSITY OF WOLLONGONG

Setyaningrum Rahmawaty Grad Dip. III Nutr (Academy of Nutrition Semarang, Indonesia) Grad Dip. IV Clin Nutr (Brawijaya University, Indonesia) MHSc. Clin Nutr (Gadjah Mada University, Indonesia)

SCHOOL OF HEALTH SCIENCES

2013

Certification

I, Setyaningrum Rahmawaty, declare that this thesis, submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy, in the School of Health Sciences, University of Wollongong, is wholly my own work unless otherwise referenced or acknowledged. The document has not been submitted for qualifications at any other academic institution.

Setyaningrum Rahmawaty

1 July 2013

“Innallooha laa yughoyyiruma biqoumin hatta yughoyyiruma bi anfusihim – verily, He will not change the good condition of a people as long as they do not change their state of goodness themselves”

Qur’an Surah Ar-Ra’d 11

iii

Dedicated to:

My country, Indonesia The memory of my Father, Mother and elder Brother

Funding sources

This study was supported by a scholarship from Directorate General of Higher Education Indonesia (DIKTI) for lecturer in universities in Indonesia and University of Muhammadiyah Surakarta Indonesia.

Small grant scheme was also provided by Metabolic Research Centre and Higher Degree Research Student Funding at the School of Health Sciences, University of Wollongong.

Acknowledgments

I would like to offer my sincere gratitude to Almighty Allah first for enabling me to undertake and successfully complete the hardest journey of my life in the form of this thesis.

If there are some moments in life when one fails to express their feelings, perhaps this is that point in my life at least. I can hardly think of proper words that can convey what I feel deep in my heart; I can only recall many ups and downs during this three a half years long venture, with friendly faces of many people around joining their hands to help me. It is not possible to list all the names and narrate how they supported me; since both time and space will fall short for this. I am grateful to all those people whose support, contribution and well wishes proved invaluable along the way. However, I feel indebted to:

- Prof. Bambang Setiadji and staffs at the University of Muhammadiyah Surakarta. Thank you very much for supporting me from the beginning to get funding from the DIKTI until finishing my PhD. - A/Prof. Barbara Meyer, thanks for being such an ‘enthusiastic and very thorough’ supervisor as well as being mom, friend and colleague; A/Prof. Karen Charlton, thanks for being a ‘smart and nice’ supervisor and; A/Prof. Philippa Lyons-Wall, thanks for being a ‘lovely’ supervisor; Dr Marijka Batterham, thanks for her ‘nice’ statistical consultant. I really enjoy learning through your different styles of guiding and supervising me to be a better scientist. - All participants in the Fish Survey and Kids’ Omega-3 Intake Study as well as a number of media persons, child care centres, mother groups, supermarkets, public offices and private primary schools in Wollongong: Sam and Adam from Mercury, thanks for the great shoot and news; Scoot from Win television; Emily from ABC Illawarra; Miss Nicole Delbridge from Illawarra Christian school and Jenny Werakso from St Brigid’s Catholic school. - Prof Paul Else, Dr. Todd Mitchell and Dr Simon Brown, thanks for sharing your knowledge about how to do cheek cell sampling and fatty acid extraction.

- Dr Jimmy Louie, thank you for his assistance in dietary analysis programming. - To Adam, Beryl, Lucy, Nola, Gai-Ellen, Sandie, Jenny and Anne, thanks a lot for your great service during my PhD at the School of Health Sciences. - Zoe, Samuel Eather, Hanieh, Sarah Norris and David, thanks so much for your kind support in data collection and working in lab (they never said ‘no/wait’ – so nice) ; Andi, Lukman, Yusuf, Dhany and Caesar many thanks for sharing your knowledge and valuable information about computer/mathematic programming. - Bothaina Bukatow, my best friend indeed!!!...I will miss walking & ‘fighting’ when exploring Wollongong and discussing our PhD progress; Josip Matesic, thanks for sharing Australia…I felt confidence to do my research here; my officemates (Dian, Li Min, Katherine, Jo, Ali, Jamila, Alisa, Khlood, Bel) - nice gossip, study, etc…wish all the best for you all friends!!! - A big thanks to my sisters, brothers, nieces, nephews for all their supports and prayers for me.

Last but not least, to my beloved late parents; my father, M Busro Busraini Burhani and my mother, Nangimah who cannot see me with this achievement. There is not even a single word that can express my deep feeling for everything they have given me in my life. May Allah shower His blessings on them and gave them a pleasant abode in ‘Jannat ul Firdous’ - Aameen. And many thanks to Najeeb’i’, your support from a thousand miles made the final stage of my PhD journey colorful.

vii

Table of contents

Page Certification ...... ii Funding sources ...... v Acknowledgements ...... vi Table of contents ...... viii Tables ...... x Figures ...... xii Abbreviations ...... xiii Publications ...... xv Abstract ...... xvii Chapter 1 - Introduction ...... 1 1.1 An overview of n-3 LCPUFA ...... 2 1.1.1 Dietary source of n-3 LCPUFA ...... 2 1.1.2 Metabolism of n-3 LCPUFA ...... 3 1.1.3 Eicosanoids metabolism...... 4 1.2 Health benefits of n-3 LCPUFA in children ...... 6 1.2.1 The role of n-3 LCPUFA on brain development and school children’s performance ...... 6 1.2.2 The role of n-3 LCPUFA in chronic-diseases prevention...... 9 1.3 Dietary n-3 LCPUFA intake and fish consumption in children’s diet ...... 12 1.3.1 International perspective on n-3 LCPUFA intake in children ...... 12 1.3.2 Dietary n-3 LCPUFA intake in Australian children...... 14 1.3.2.1 Dietary recommendations for n-3 LCPUFA intake for Australian children...... 14 1.3.2.2 Situational analysis: dietary n-3 LCPUFA intake in Australian children ...... 16 1.4 Factors that influence fish/or seafood consumption...... 17 1.5 Foods enriched with n-3 LCPUFA – an alternative source of n-3 LCPUFA intake and their potential health benefits ...... 18 1.6 Methods to improve dietary intake in children ...... 25 1.6.1 Intervention Mapping – a stepwise approach for designing behaviour changes development ...... 24 1.6.2 Parent’s influence to children’ eating habit ...... 26 1.6.3 Dietary modelling as a strategy to meet dietary recommendation...... 26 1.6.4 Food pattern analysis ...... 27 1.7 Methodological issues associated with assessing fish and n-3 LCPUFA intake in children ...... 28 1.7.1 Food frequency questionnaire as a dietary assessment method for food periodically consumed...... 29

viii

1.7.1.1 Pros and cons of FFQ compared to other dietary instruments ...... 29 1.7.1.2 Validation of FFQ...... 30 1.7.2 Development of an FFQ for children...... 31 1.8 Summary of evidence and gaps in the literature ...... 32 1.9 Aims and hypotheses ...... 33 1.10 Thesis structure ...... 35 1.11 References ...... 36

Chapter 2 - Paper 1: Dietary intake and food sources of EPA, DPA and DHA in Australian children ...... 58

Chapter 3 - Paper 2: Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children ...... 79

Chapter 4 - Paper 3: Effect of replacement of bread, egg, milk and yogurt with n-3 equivalent n-3 enriched foods on n-3 LCPUFA intake of Australian children ...... 99

Chapter 5 - Paper 4: Food patterns of Australian children aged 9-13 years in relation to n-3 long chain polyunsaturated (n-3 LCPUFA) intake ...... 127

Chapter 6 - Paper 5: Development and validation of a food frequency questionnaire to assess n -3 LCPUFA intake in Australian children aged 9-13 years ...... 148

Chapter 7 - Summary ...... 169 7.1 Contextualising the thesis...... 170 7.2 Core of thesis findings ...... 171 7.3 Future direction and recommendations ...... 173

Appendices ...... 177 (a) Fish and seafood survey questionnaire ...... 178 (b) An n-3 LCPUFA FFQ for Australian Children aged 9-13 years ...... 184 (c) Food Model booklet ...... 206

Tables

Page

Global recommendation for fish and n-3 LCPUFA intake in children (mg/d) ...... 12

Mean intake of EPA, DPA and DHA of different population studies of children ...... 13

The nutrient reference values for n-3 LCPUFA for Australian children ...... 16

Average daily intake of long chain n-3 PUFA for all children by age group and sex (mg/d) ...... 65

Average daily intake of EPA, DPA, DHA and total LC n-3 PUFA for all children by fish eater status, adjusted and unadjusted for energy intake ...... 66

The consumption (g/d) of major food sources of long chain n-3 PUFA (meat, egg and fish) by fish eater status in relation to percentage of children meeting SDT ...... 67

Intakes of EPA, DPA, DHA and total LC n-3 PUFA in correlation with the intakes of some food groups (Kendall’s tau-b correlation coefficients) ...... 68

Percentage contribution of food groups to EPA, DPA, DHA and total long chain (LC) n-3 PUFA for all children (n 4,487) aged 2-16 y ...... 68

Characteristic of participants (n =262) ...... 85

Frequency for consuming fresh or frozen fish or seafood products (% of total participants, n = 262) ...... 86

Frequency for consuming canned fish or seafood products (% of total participants, n = 262) ...... 86

Omega-3 LCPUFA content in n-3 LCPUFA enriched bread, egg, milk and yogurt .... 106

Intake of bread, egg, milk and yogurt in consumers ...... 106

Modelling of usual daily intake of EPA, DPA, DHA and total n-3 LCPUFA before and after replacement with n-3 enriched foods for all children (n = 4487) ...... 107

Modelling of usual daily intake of EPA, DPA, DHA and total n-3 LCPUFA before and after replacement with n-3 enriched foods (n = 3554) in non-fish eater ...... 109

Modelling of usual daily intake of EPA, DPA, DHA and total n-3 LCPUFA before and after replacement with n-3 enriched foods (n = 933) in fish eater ...... 110

Modelling of the average intake of EPA, DPA, DHA and total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods (n = 4487) .. 118

Modelling of the average intake of EPA, DPA, DHA and total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods (n = 3554) in non- fish eater ...... 119

Modelling of the average intake of EPA, DPA, DHA and total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods (n = 933) in fish eater ...... 120

Food grouping used in this study of Australian children (age = 9-13 years, n = 1,110) ...... 133

Food patterns of Australian children by gender (age = 9-13 years, n = 1,110) ...... 137

Characteristics of Australian children within each food pattern group, by gender (age = 9-13 years, n = 1,110) ...... 138

Relationship between food patterns and n-3 LCPUFA intake status in Australian children (age = 9-13 years, n = 1,110) - Discriminant analysis (test t of equality of group means) ...... 139

Food items assessed in the n-3 LCPUFA FFQ ...... 156

Comparison of EPA, DPA, DHA and total n-3 LCPUFA obtained from FFQ and the 7d food diary ...... 157

Validity correlation between FFQ and the average 7d food diary ...... 157

Log transformed limit of agreement between FFQ and the average 7d food diary...... 158

Figures

Page

Metabolism of n-3 LCPUFA and eicosanoids ...... 6

Overview of thesis studies in relation to the central hypothesis...... 35

Factors that encourage (positive) and prevent (negative) for consuming fresh/frozen fish/seafood (% of respondents) ...... 88

Factors that encourage (positive) and prevent (negative) for consuming canned fish/seafood (% of respondents) ...... 89

Frequency for consuming omega-3 enriched foods (% of total respondents, n = 262) ...... 90

Factors that encourage (positive) and prevent (negative) for consuming omega-3 enriched foods (% of total respondents, n = 262) ...... 90

The changes distribution of usual daily dietary intake of total n-3 LCPUFA of the Australian children’s diet (n = 4487) before and after replacing bread, milk, yogurt and egg with enriched n-3 for these foods ...... 111

Modelling intake of total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods from (n = 4487) in relation to the nutrient reference values ...... 121

Bland-Altman plot: comparison of the agreement of dietary EPA intakes determined using food frequency questionnaire and 7d food diary, after log transformation ...... 158

Bland-Altman plot: comparison of the agreement of dietary DPA intakes determined using food frequency questionnaire and 7d food diary, after log transformation ...... 159

Bland-Altman plot: comparison of the aggreement of dietary DHA intakes determined using food frequency questionnaire and 7d food diary, after log transformation ...... 159

Bland-Altman plot: comparison of the aggreement of dietary total n-3 LCPUFA intakes determined using food frequency questionnaire and 7d food diary, after log transformation ...... 160

xii

Abbreviations

AA arachidonic acid ADHD attention-deficit hyperactivity disorder AI adequate intake ALA alpha-linolenic acid (C18:3n-3) ANCNPA Australian National Children’s Nutrition and Physical Activity ANOVA analysis of variance AUSNUT Australian nutrient composition database BMI body mass index cAMP cyclic adenosine monophosphate CAPI computer assisted personal interview CATI computer assisted telephone interview CI confidence interval CRP C-reactive protein CVD cardiovascular disease COX cyclo-oxygenases CSIRO Commonwealth Scientific and Industrial Research Organization d day DCD development coordination disorders DET Department of Education and Training DHA docosahexaenoic acid (C22:6n-3) DPA docosapentaenoic acid (C22:5n-3) EAR estimated average requirement EFA essential fatty acid EPA eicosapentaenoic acid (C20:5n-3) FA fatty acid FAE-2 fatty acid elongase-2 FAE-5 fatty acid elongase-5 FAO/WHO Food and Agriculture Organization of the United Nations and the World Health Organisation FFQ food frequency questionnaire FMD flow-mediated dilation FRDC Fisheries Research and Development Corporation FSANZ Food standards Australia and New Zealand g grams GM genetic modification h hour HDL high density lipoprotein HNF-4 hepatic nuclear factor-4 ICAM inter-cellular adhesion molecules ID identity IL interleukin IQ intelligence quotient IOM Institute of Medicine xiii

IQR inter quartile range Kcal Kilo calorie KMO Kaiser-Meyer-Olkin LA linoleic acid LDL low density lipoprotein LOV lacto-ovo-vegetarian LOX lipoxygenases LTs leukotrienes LXR lier-X factor α mg miligrams mg/d/MJ milligram/day/mega joule MATLAB MathWorks MTO microencapsulated tuna oil n-3 LCPUFA omega-3 long chain polyunsaturated fatty acid NHFA National Health Foundation of Australia NHMRC National Health and Medical Research Council NNS National Nutrition Survey NSW New South Wales PCA Principle-component analysis PL phospholipids PGs prostaglandins PPARs peroxisome proliferator-activated receptors PPVT Peabody Picture Vocabulary Test PKU phenylketonuria PUFA polyunsaturated fatty acid RBC red blood cells RDA recommended dietary allowance RDD random digit dialling Rvs resolving SCD stearoyl-CoA desaturase SERAP State Education Research Approval Process SREBP-1 sterol regulatory element binding protein-1 SD standard deviation SDT suggested dietary target SPSS statistical package for social sciences TAG triacyl glycerol TG triglyceride TNF tumor necrosis factor TX tromboxane VC validity coefficient

xiv

Publications

This thesis is based on the following papers (peer-reviewed publications) and they are referred to in the text by their roman numerals.

Paper 1 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Dietary intake and food sources of omega-3 long chain EPA, DPA and DHA of Australian children. Lipids 48: 869-877.

Paper 2 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Nutrition and Dietetics. DOI: 10.1111/1747-0080.12022.

Paper 3 Rahmawaty S, Lyons-Wall P, Charlton K, Meyer BJ (2013). Effect of replacement of bread, egg, milk and yogurt with n-3 enriched for these foods on n-3 LCPUFA intake of Australian children. Nutrition, submitted.

Paper 4 Rahmawaty S, Lyons-Wall P, Charlton K, Batterham M, Meyer BJ (2013). Food patterns of Australian children aged 9-13 years in relation to omega-3 long chain polyunsaturated (n-3 LCPUFA) intake. Nutrition, http://dx.doi.org/10.1016/j.nut.2013.07.014

Paper 5 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Development and validation of a food frequency questionnaire to assess n-3 LCPUFA intake in Australian children aged 9-13 years. Public Health Nutrition, submitted.

Abstract Publications in support of this thesis

Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2010) Dietary intake and food sources of omega-3 long chain EPA, DPA and DHA of Australian children. Proceedings of the Nutrition Society of Australia 34, 6.

Rahmawaty S, LyonsWall P, Charlton K, Meyer BJ (2012) Dietary modelling of omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA): effect of replacement of bread, egg, milk and yogurt with n-3 enriched for these foods on n-3 LCPUFA intake of Australian children. The 2nd International Food Studies at the University Illinois at Urbana Champaign USA, 4 – 5 October 2012

Rahmawaty S, LyonsWall P, Charlton K, Meyer BJ (2012) Dietary modelling of omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA). Proceedings of the Nutrition Society of Australia 36, 45.

Rahmawaty S, Charton K, LyonsWall P, Meyer BJ (2012) Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Proceedings of the Nutrition Society of Australia 36, 70.

Presentations in support of this thesis

1. The 34th Nutrition Society of Australia, Perth 2010 (oral presentation) 2. Nutrition Australia NSW Division, Annual General Meeting (oral presentation) 3. The 2nd International Food Studies at the University Illinois at Urbana Champaign USA, 4 – 5 October 2012 (oral presentation) 4. The 36th Nutrition Society of Australia, Wollongong 2012 (oral and poster presentation)

xvi

Abstract

An accurate dietary assessment of n-3 LCPUFA intake is necessary to monitor the possibility of the emergence of nutrition issues related to the consumption of these fats. In Australia, however, limited information is available on dietary assessment of EPA, DPA and DHA intake in the population of children.

The central aim of this thesis is to assess current intakes of EPA, DPA and DHA, and options that may potentially be used to modify intake of these fatty acids in Australian children’s diet.

The research designs used in this thesis are as follows: (1) Secondary analysis of n-3 LCPUFA intakes of the 2007 Australian National Children’s Nutrition and Physical Activity Survey; (2) Survey of barriers and promoting factors to the consumption of fish and foods enriched with n-3 in Australian families with young children; (3) Dietary modelling of n-3 LCPUFA; (4) Food patterns analysis in relation to n-3 LCPUFA; and (5) Development and validation of an n-3 LCPUFA FFQ for Australian children.

The main results were as follows: (1) Energy-adjusted intakes of EPA, DPA and DHA in children who ate fish were 7.5, 2 and 16-fold higher, respectively (p < 0.001) compared to those who did not eat fish during the two days of the survey. Fish and seafood products were the largest contributors to DHA (76 %) and EPA (59 %) intake, while meat, poultry and game products contributed to 56 % DPA. (2) Perceived health benefits, and the influence of media and health professionals were identified as the primary motivators for consumption of fish/seafood, and foods enriched with n-3 LCPUFA. Among families who consumed fish, taste was valued as having a major positive influence, as well as preferences of individual family members, but the latter were perceived as an obstacle in non-fish consumers. Price was the main barrier to consumption of fresh, but not canned, fish and n-3 enriched foods, in both those that did and did not consume these foods.

xvii

(3) Replacement of four commonly consumed food items with n-3 enriched for these foods would result in substantially improved n-3 intakes in Australian children, especially in non-fish eaters, without major changes to their current food habits. (4) Dietary patterns associated with a high consumption of vegetables in boys and take-away fish in girls were likely to positively influence dietary n-3 LCPUFA intake in Australian children. (5) The n-3 LCPUFA FFQ developed in this study provided an acceptably valid estimate of dietary n-3 LCPUFA intake of Australian children aged 9-13 years, when compared to intake obtained by a 7-day food diary.

The results of this thesis provide a number of resources that could be used to develop and monitor nutrition intervention programs designed to increase n-3 LCPUFA intakes of Australian children.

Keywords: Australia, children, food analysis, n-3 LCPUFA, monitoring

xviii

Chapter 1. Introduction

Chapter 1

I n t r o d u c t i o n

Chapter 1. Introduction

This chapter focuses on dietary omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) eicosapentaenoic acid (EPA, C20:5n-3), docosapentaenoic acid (DPA, C22:5n-3) and docosahexaeoic acid (DHA, C22:6n-3) intake in the context of in-depth nutrient analyses to provide an understanding of aspects which could potentially be used in designing and monitoring intakes and nutrition intervention program. The gaps in the current literature are identified, and methodological issues associated with assessment of dietary n-3LCPUFA intake highlighted.

Limited information is available on levels of n-3 LCPUFA intake in Australian children. Secondary analyses of the most recent national dietary survey data, the 2007 National Children’s Nutrition and Physical Activity (Children’s Survey) were undertaken, and the findings used as a corner stone to develop the sequential next studies in this thesis.

1.1 An overview of n-3 LCPUFA

Since they were discovered in 19291, n-3 LCPUFA including EPA, DPA and DHA have continuously received increased attention, particularly after the observation that Greenland’s Eskimos have low incidence of cardiovascular disease (CVD) in the setting of a diet rich in fatty fish2,3,4. Recently, evidence of the causality showing that n-3 LCPUFA decrease a death from CVD5,6. Furthermore, CVD is the leading cause of death in Australia, whereby 90% of Australian adults have at least one modifiable CVD risk factor and 25% have three or more modifiable risk factors7. The prevalence of cardiovascular disease (CVD) risk factors in Australian children has been reported to rise with increasing age8,9.

The n-3 LCPUFAs are unsaturated fatty acids which have double bond in the acyl chain, whereby the last double bond is located three carbons from the methyl (omega or nth) end of the fatty acids10. These fatty acids (FAs) are a series of metabolic products of short chain n-3 called α-linolenic acid (ALA, C18:3n-3), an essential fatty acid (EFA) which cannot be synthesized by mammals but can be synthesized by plants11.

1.1.1 Dietary source of n-3 LCPUFA

Omega-3 LCPUFA are predominantly provided pre-formed from dietary sources, and minimal ability for de novo synthesis in liver from elongation of the shorter chain n-3 2

Chapter 1. Introduction

FA, ALA to DHA12,13,14 found in nut, soy, flaxseed, canola oil and rapeseed oil. Omega- 3 LCPUFA are initially produced by small aquatic plans, marine micro-algae that are consumed at the bottom of the marine food chain, accumulated in the fish/seafood chain15. The richest food sources of n-3 LCPUFA are marine fish, particularly oily fish such as mackerel, herring, anchovies, sardines, tuna, salmon and oyster15. Lower amounts are found in meat16, especially lean meat, eggs and milk17. Food enriched with n-3 LCPUFA e.g. bread, eggs, milk products, margarine, spreads and infant formula fortified with n-3 also provide a small to moderate amount of n-3 LCPUFA18. Likewise, meat and meat products may be enriched with n-3 LCPUFA via the animals’ diet19,20. Other potential source of n-3 LCPUFA is genetic modification (GM) by transferring the genes responsible for the production of n-3 LCPUFA to land based plants21,22,15.

1.1.1.1 Metabolism of n-3 LCPUFA

Once ALA are obtained from diet, they are converted to n-3 LCPUFA series by a series of interchanging desaturation and elongation reaction involving Δ6 and Δ5 desaturase and elongase enzymes respectively23-26. The desaturase inserts double bonds into the molecules and the elongase adds 2-carbon units to the fatty acids. The conversion of ALA to the n-3 LCPUFA series occurs primarily in the liver in the endoplasmic reticulum. There is competition between ALA and linoleic acid (LA, C18:2n-6), in the use of Δ6 and Δ5 enzymes to produce their long chain PUFA (Figure 1.1). The dietary intake level of these fatty acids influences the physiological actions of their active metabolites. Diets high in ALA increase the rate of ALA oxidation, limiting its accumulation in plasma and reducing its conversion rate to EPA and DHA27. It has been argued that an increase in ALA intake can reduce the eicosanoids or metabolites of the n-6 pathway, particularly for AA28,29 that has potent pro-inflammatory effects30. In contrast, diets high in LA reduce the conversion of ALA to its long-chain derivatives by 40 %, with a net reduction in n-3 LCPUFA accumulation of 70 %12.

In the human body, n-3 LCPUFA are oxidized in the mitochondria. They are integrated structurally and functionally through phospholipid molecules31 and incorporated into the phospholipid bilayer as the major components of cellular membranes.

Chapter 1. Introduction

Omega-3 LCPUFA have been reported to be minimally synthesized de novo from elongation of ALA12,14,32. There are studies of patients with n-3 deficiency that have reported an increase in EPA and DHA plasma concentration following ALA intervention33,34. Additionally, stable concentrations of DHA in plasma have been reported in vegans, a group that consumes no animal products35,36. However, most isotope-tracer studies in healthy subjects are consistent with a limited conversion of ALA to DHA37-39. In a study of children aged 0-18 years with disorders of amino acid metabolism who consume protein-restricted diets (and hence do not consume preformed n-3 LCPUFA e.g. fish or seafood, egg and meat products), their tissue DHA concentrations were 30 % lower than healthy children, whilst concentrations of arachidonic acid (AA) did not differ between the groups40. This evidence suggests that there is no problem with elongation and desaturation enzymes because LA is converted to AA. However, due to high LA levels, the ability to convert ALA to DHA is reduced40. A cross-sectional study of children with allergies, who eliminate food groups such as milk, egg, fish and vegetables, reports that those children have approximately two-fold lower levels of EPA and DHA in plasma FA compared to healthy children, whilst concentrations of AA and ALA do not differ41. These studies suggest that pre-formed n-3 LCPUFA needs to be consumed to fulfil physiological requirements, and that conversion of ALA to n-3 LCPUFA is not adequate for optimal health.

Omega-3 DPA is an elongated metabolite of EPA and is an intermediary in the conversion of EPA and DHA42. The conversion of EPA to DPA is mediated by the enzymes fatty acid elongase-2 (FAE-2) and FAE-543. EPA and DHA are actively interchanged in the endothelial cells, resulting in the accumulation of DPAn-344. The concentration of DPA in tissue depends on the balance from EPA and its conversion to DHA32.

1.1.2 Eicosanoids metabolism

EPA can be metabolized via cyclo-oxygenases (COX) or lipoxygenases (5-LOX) metabolic pathways leading to the formation of hormone-like substances called eicosanoids. Eicosanoids are active metabolites of lipid-based molecule that operated as mediators of inflammation and immunity45. Eicosanoids are produced as a result of a stimulus. EPA is the precursor eicosanoids for 3-series prostaglandins (PGs), 4

Chapter 1. Introduction

46 thromboxane A3 (TXA3) and 5-series leukotrienes (LTs) . Other eicosanoids, E-series resolvins (Rs, resolution phase interaction products) are also produced from EPA and mediated by COX-2. DHA is converted by COX-2 and LOX to docosanoids called D- series resolvins and protectins (neuroprotectins D1)47-49 such as docosatrienes and maresins, a class of macrophage mediators50,51. The eicosanoids and docosanoids are the signalling molecules in the body that play crucial roles in the regulation of broad physiological actions e.g. modulation of inflammation, blood pressure regulation, platelet aggregation, blood clotting and blood lipid profiles10,52.

Mitochondria

Figure 1.1: Metabolism of n-3 LCPUFA and eicosanoids, adopted from: Ratnayake & Galli, 200910; SanGiovanni & Chew 200545 with some modifications 5

Chapter 1. Introduction

AA is the substrate eicosanoids for 2-series PGs, TXs and 4-series LTs that produce proinflammatory cytokines, a key process in the inflammatory response45. Leukotriene B4 produced by AA is associated with proinflammatory cytokine called tumor necrosis factor (TNF) alpha53 that mediate production of a number of potent proinflammatory and immunoregulatory cytokines54, and they all associated with vascular leakage45.

EPA and DHA demonstrate potential effect in modulating production of AA- derived eicosanoids by inhibition/inactivation of enzymes responsible in the production of the eicosanoids and action of eicosanoids45. A recent review on the biological effects of DPA showed that n-3 DPA is a more effective platelet aggregation inhibitor than EPA and DHA55. An in vitro study demonstrated that, the ability of DPA to possess endothelial cell migration was 10-fold higher than EPA, which is important in wound- healing processes56. An in vivo study has reported that n-3 DPA reduces the fatty acid synthase and malic enzyme activity levels in n-3 DPA-supplemented mice and that these effects were stronger than for EPA-supplemented mice. Another recent in vivo study has reported that n-3 DPA may have a role in attenuating age-related decrease in spatial learning and long-term potentiation55.

1.2 Health benefits of n-3 LCPUFA in children

1.2.1 The role of n-3 LCPUFA on brain development and school children’s performance

DHA is a crucial element in the nervous system, which is responsible for development of the sensory, perceptual, cognitive and motor neural system during the brain growth spurt45,57. It is the most fluidizing compound in cell membrane58, which is fundamental for growing membranes59. The retina, functionally an extension of the brain, is highly enriched in DHA, especially in retinal photoreceptor outer segment disc membranes60. DHA is highly accumulated in the brain and retinal photoreceptor cells, forming 10-20 % and > 60 % of total fatty acid (FA) composition in the brain and retina, respectively61,62. During the third trimester of gestation (week 26-40), there is a rapid increase (~14.6 mg/week) of accretion of DHA in the foetus63,64. The accretion of DHA in the brain continues up to 2 years of age and progressively rises in the cerebral cortex until 18 years of age65,66 and even life-long67. A study has compared brain tissue

Chapter 1. Introduction

obtained from individuals ranging from fetal age to 82 years to assess percentages of AA and DHA methyl esters in cerebral cortex ethanolamine glycerophospholipids67. Concentrations in the one month old human infant were roughly equal or 1:1 (16.5 % AA and 16.1 % DHA) but by the 82nd year the percentage of DHA had more than doubled and increased in proportion to AA, to approximate a ratio of 1:4 (10.3 % AA and 33.9 % DHA)67. Concentration of DHA in plasma phospholipids (PL) and red blood cells (RBC) rapidly falls by ~50 % within 4 months after birth without an exogenous source of DHA, but is maintained by consumption of human breast milk or DHA- fortified formula feeding68.

The n-3 LCPUFA are involved in various neuronal processes, ranging from regulation of gene transcription to effects on cellular signalling processes69, and protection against the pathogenesis of retinal diseases45. Nutrient deficiency during the brain growth spurt, which is between the last trimester of gestational and the first two years of childhood can damage brain function57.

Epidemiological studies have reported that fish consumption in school children is significantly associated with higher school grades70, better cognitive performance71, higher intelligence quotient (IQ)72 and more positive mental states (beliefs and intentions)73. Swedish children (age 16 y, n = 9448) who ate fish at least once or twice a week had a mean grade of 19 and 30 points higher, respectively, than children who consumed fish less than once a week70. A strong positive association between fish consumption and global intelligence score, as well as performance on verbal and visuospatial tests, has been reported in a 3-year follow-up study (n = 3972)71. A longitudinal study of New Zealand Children (n = 591) has also reported a significant positive association between weekly fish consumption and cognitive ability. The total intelligent quotient (IQ) score of children who ate fish at least once a week was 3.64 points higher compared to those children who did not eat fish72.

An increase in consumption of food sources containing n-3 LCPUFA e.g. fish, seafood70,71 and foods fortified with n-374,75 has been found to be associated with an improvement of cognitive performance in school children. A randomized placebo- controlled, double-blind study showed a significant positive association (r2=0.14, p=0.018) between blood levels of DHA and higher scores on the Peabody Picture Vocabulary Test (PPVT) in healthy preschool children74. A six month randomly 7

Chapter 1. Introduction

assigned experimental and control group intervention study has reported that consumption of bread spread with fish-flour suppling 892 mg DHA per week, equivalent to 125 mg or 2 portions of medium fat fish per week significantly increased EPA and DHA, which had a positive impact on verbal learning and memory of African children (age 7-9 y)75. Similar findings have been reported in an intervention study of Indonesian and Australian primary school children who received supplementation with multiple micronutrients, as well as EPA plus DHA76. Improvement of biochemical markers (e.g. anaemia) has also been reported in a double blind-randomized controlled trial study in primary school Indian children after supplementation with multiple micronutrients and DHA77. These studies provide evidence that consumption of marine- enriched food not only supported optimal health in school-aged children but also 74,75 showed improvements in their academic achievement .

Numerous experimental studies using animals, rodents and primates showed that n-3 deficiency results in deficits in retinal structure, visual acuity development and cognitive performance78-80. A number of evidence in animal studies reported that deficiency of n-3 PUFA intakes during the developmental period is associated with a decline in neurons that negatively affected the performance of rats. Inadequate n-3 PUFA during conception resulted in depletion in both the number and function of brain dopamine source neurons in the rat hippocampus81. Depletion of ALA intake among dams and offspring during the peri-natal, post-weaning and post-pubertal stages of development produced a graded 10-60 % decline in forebrain serotonin and serotonin transporter content82. Animal studies of n-3 deficiency during infancy have reported outcomes of poor learning and memory performance in a variety tests, including the Morris Water Maze83,84, olfactory-based learning and memory tasks involved mainly in complex learning85,86, spatial tasks and olfactory-cued reversal learning tasks87 as well as sensory deficits in the form of visual problems88,89. This evidence suggests that sufficient intake of n-3 LCPUFA is crucial for supporting normal brain, cognitive and retina development, and protection against neuro-psychiatric disorders. Moreover, the brain keeps producing neurons, even into adulthood90. Research in primates and rodents has found that the brain cortex undergoes highly active synaptic turnover throughout life91. This suggests that sufficient intake of n-3 LCPUFA is fundamental, not only during the growth spurt period, but also across the life span.

Chapter 1. Introduction

A number of hypotheses have been proposed to explain the poor learning of individuals with n-3 PUFA deficiency. The n-3 PUFA deficiency results in a significantly decreased number of neurons in the hippocampus, hypothalamus and cortex, brain areas that mediate spatial and serial learning92, cerebral catecholamines93, glucose transport capacity and glucose utilization in the brain94, cyclic adenosine monophosphate (cAMP) level in the hippocampus95 and capacity of phospholipid synthesis in the brain and hypothalamus96,97. The cAMP is a second messenger that is important in many biological processes. Any changes in the levels of catecholamines, glucose, cAMP and phospholipids can pose learning deficits.

1.2.2 The role of n-3 LCPUFA in chronic-diseases prevention

Potential mechanisms related to the role of n-3 LCPUFA in reduction of inflammatory responses have been reviewed by Larsson et al.29. It was concluded that high intakes of n-3 LCPUFA can inhibit the desaturation and elongation of LA to AA, and suppress the biosynthesis of the AA-derived eicosanoids that have pro-inflammatory effects98-100. An increase in the formation of EPA-derived eicosanoids also inhibits the action of the AA- derived eicosanoids.

Prostaglandins and thromboxane from EPA act as vasodilators and anti- aggregators49. The resolvins produced from EPA and DHA appear to have potent anti- inflammatory activities through inhibition of leukocyte activation and inflammatory mediator synthesis101. The DHA-derived neuroprotectin D1 is formed in retinal pigment epithelial cells when they are confronted with oxidative stress, as well as in the brain during experimental stroke and are found in the human brain of people with Alzheimer ’s disease, as well as in human brain cells in culture102. Neuroprotectin D1 demonstrates potent anti-inflammatory and neuroprotective bioactivity through down-regulating brain ischemia reperfusion induced leukocyte infiltration, proinflammatory signalling and infarct size, and also inhibits cytokine-mediated cyclooxygenase-2 expression102. Resolvins and docosatrienes show potent anti-inflammatory and immunoregulatory actions50.

The n-3 LCPUFA modulate signalling pathways which are involved in lipid metabolism. A review of evidence in rodents showed that EPA and DHA activate peroxisome proliferator-activated receptors (PPARα) and simultaneously inhibit sterol

Chapter 1. Introduction

regulatory element binding protein-1 (SREBP-1), which stimulates β-oxidation and inhibits FA synthesis, and consequently decreases serum triacylglycerol103. PPARα is a major regulator of lipid metabolism in the liver104. Activation of PPARα promotes uptake, utilization, and catabolism of fatty acids by upregulation of genes involved in fatty acid transport, fatty binding and activation, and peroxisomal and mitochondrial fatty acid β-oxidation104. SREBP-1 is a lipid synthetic transcription factor especially for cholesterol and fatty acid synthesis105. Limited studies have been conducted in humans.

The progression of risk factors for chronic diseases such as CVD that are associated with age can be either delayed or accelerated as a function of lifestyle choice106. Dietary habits in childhood influence the development of an individual's CVD risk profile in later life107, whilst, early intake of n-3 LCPUFA has shown beneficial effects regarding the prevention of CVD in later life108,109. Several randomized double- blind studies in children aged between 2-18 years have demonstrated beneficial outcomes for a lower risk of development of chronic diseases61,109-111and a better CVD risk factor profiles in children following n-3 LCPUFA supplementation112-114. Low concentrations of DHA have been found to be associated with increased concentrations of C-reactive protein (CRP), a sensitive predictor of CVD risk115, in overweight teenagers116. Conversely, an increased intake of n-3 LCPUFA has been shown to modulate vascular functions and inflammatory markers, including lymphocytes, monocytes and level of TNF-α, interleukin (IL)-1β and IL-6, in obese adolescents112. A low n-3 index, the sum of erythrocyte EPA and DHA, expressed as a percentage (weight/weight) of total fatty acid concentrations in red blood cell membranes (but not in plasma)117, has also been found in obese school age children and is associated with insulin resistance118. The n-3 index is considered as a potential biomarker that can identify persons at risk for coronary heart disease in adults119,120. A value of less than 4 % is categorized as high risk while > 8 % is categorized as low risk119. It has been reported that higher intake of EPA and DHA through fish consumption is associated with higher n-3 index in healthy adolescents and reduced CVD factors in boys121.

Decrease in DHA levels was associated with increased stearoyl-CoA desaturase (SCD), a desaturating enzyme that modulates fatty acid composition and might contribute to development of metabolic syndrome in obese children122. A reduction in biomarkers of CVD risk namely E-selectin and inter-cellular adhesion molecules

Chapter 1. Introduction

(ICAM)-1 has also been observed in healthy children aged 8-14 years following n-3 LCPUFA supplementation123. E-selectin and ICAM-1 are members of cell adhesion molecules that are involved in the inflammatory development to endothelial damage, such as seen in CVD124,125. Additionally, improvement in endothelium-dependent flow- mediated dilation (FMD) of the brachial artery after DHA supplementation (1.2 g/d for 6 weeks) has also been reported in hyperlipidemic children at risk for early heart disease126. Supplementation of 900 mg/d n-3 LCPUFA for a month has been reported to have synergistically reduced insulin resistance together with weight loss in obese at-risk children. The mechanistic effect may be through an increased production of adiponectin and a decrease in inflammatory biomarkers such as TNF- α127.

An intervention study in healthy children (8-14 y) has reported that consumption of 600 mL/d milk enriched with fish oil, oleic acid, minerals and vitamins (Puleva Max containing fish oil, EPA = 10 mg/100mL and DHA = 20 mg/100 mL, n-3 = 35 mg/100mL; oleic acid; carbohydrates (sugar and honey); vitamins: A as retinyl ester, B1, B2, B3, pantothenic acid, B6, biotin, folic acid, B12, C, D, and E; minerals: calcium, phosphorus, zinc) that was low in saturated fatty acids reduced incidence of endothelial cell activation, a CVD risk biomarker123.

The body of evidence cited in this section suggests that a sufficient intake of n-3 LCPUFA during childhood and adolescence is not only necessary for supporting normal cognitive development during childhood but also appears to reduce risk factor for CVD. Atherosclerosis, the cholesterol build up in arteries, is developed over time. Reduction in risk factors for CVD reduces the progression of atherosclerosis.

Moreover, intervention studies in adults have reported that EPA and DHA supplementation reduced decrease risk of mortality and possibly non-fatal incidence of coronary heart disease3-6,128-133. Furthermore, a review of nine controlled intervention studies in healthy adults and individuals with CVD risk factors, as well as CVD patients, has reported that foods enriched with EPA and DHA reduced blood lipids, particularly cholesterol, LDL-cholesterol and triglycerides114.

Chapter 1. Introduction

1.3 Dietary n-3 LCPUFA intake and fish consumption in children’s diets

1.3.1 International perspective on n-3 LCPUFA intake in children

Dietary recommendations regarding optimal n-3 LCPUFA intake for children varies across countries, due to differences in the definition of optimal nutrition, for example whether preventing deficiency or promoting good health134. A review of the available evidence for establishing dietary recommendations for n-3 LCPUFA in children suggests that dietary recommendations for children should be consistent with advice for adult populations, that is an intake of DHA and/or EPA equivalent to 1 to 2 fatty fish meals per week or approximately 500 mg of EPA and DHA per day in order to reduce cardiovascular diseases risk135. Within the context of recommended n-3LCPUFA intakes, dietary advice for the prevention of nutrition-related chronic diseases include targets for type of fat intake. Polyunsaturated n-3 should provide 1-2 % of energy in the diets of children older than 2 years and the n-6/n-3 ratio should ideally be in the range of 5/1 to 10/1135.

Table 1.1: Global recommendation for fish and n-3 LCPUFA intake in children (mg/d)136

There is general concern that dietary n-3 LCPUFA intakes of children in many developed and developing countries are reported to be much lower than recommended desirable intakes (Table 1.2). An exception is in the case of Japan, where the mean fish 12

Chapter 1. Introduction

Table 1.2: Mean intake of EPA, DPA and DHA of different population studies of children

Country Number of Method Definition of total Fish/ EPA DPA DHA Total n-3 LCPUFA children and age n-3 LCPUFA seafood g/d mg/d % of E mg/d % of E mg/d % of E mg/d % of E Australia17 383 (2-3 y) 24h recalls Sum of EPA, DPA n/a 10 1.6* 5 0.8* 24 3.8* 40 6.3* 799 (4-7 y) and DHA n/a 19 2.6* 10 1.3* 47 6.3* 76 10* 739 (8-11 y) n/a 30 3.3* 17 1.9* 60 6.7* 106 12* 653 (12-15 y) n/a 32 3.2* 22 2.2* 63 6.2* 117 12* 433 (16-18 y) n/a 41 3.7* 20 1.8* 77 6.9* 138 12* US137 962 (6-11 y) 1d x 24h recall n/a n/a 10 (0) n/a 10 (0) n/a 40 (10) n/a n/a n/a 2208 (12-19 y) n/a 20 (0) n/a 10 (0) n/a 50 (10) n/a n/a n/a Australia16 1921 (2-11 y) 24h recalls Sum of EPA, DPA n/a 32 n/a 32 n/a 46 n/a 110 n/a 1086 (12-18 y) and DHA n/a 57 n/a 63 n/a 75 n/a 195 n/a Chinese138 196 (1-5 y) 3 x 24h recall n/a n/a n/a n/a n/a n/a 30± 40 § n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a Belgium142 661 (2.5-6.5 y) Parentally reported Sum of LNA, EPA, 8.6‡ 25 n/a 10 n/a 47 n/a 941 n/a 3d EDR DPA and DHA n/a (1-21)§§ n/a (2-10)§§ n/a (5-46)§§ n/a (730-1100)§§ n/a German141 1024 (2-18 y) 3d weighed records Sum of EPA and 5.4-13.9†† n/a n/a n/a n/a n/a n/a 42-141 0.04-0.05% (yearly) DHA n/a n/a n/a n/a n/a n/a n/a n/a n/a

139 Guatemalan 219 (8-12 y, LSES) Single pictorial 24h n/a n/a 8 (2)§† n/a n/a n/a n/a 32 (2) §† n/a n/a 230 (8-12 y, HSES) record n/a n/a n/a n/a n/a n/a n/a 10 (2)§† 32 (2) §†

EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; LNA, linolenic acid; LSES, Low social economy status; HSES, high social economy status; n/a, not available; *, mg/d/energy (MJ); §, Mean ± standard deviation; §§, Mean (inter quartile range); §†, Mean ± SEM; ∞, Mean ± median; ††, The lower and the higher mean of all age groups; ‡ Seafood

13 Chapter 1. Introduction

consumption of children (12-15 years) is 17.9 ± 8.9 g/1000 Kcal, and mean energy intake is 2206 ± 595 Kcal/d, which corresponds to an EPA+DHA intake of 0.17 ± 0.1 % of total energy140.

Intakes of n-3 LCPUFA (EPA plus DHA) in children that did not consume fish was at least twice as high as in those that did consume fish141. Fish consumption, even if low, is a major contributor to total n-3 LCPUFA intake. A study in Belgium has reported that fish and seafood were the highest contributors to EPA, DPA and DHA intake (53.5, 42.8 and 48.1 %, respectively); however, only 31 % of children in this study consumed fish. The mean of fish was 8.6 g/d in total children, and it was approximately 3-fold higher (27.4 g/d) in the seafood consumer142. It can be concluded that low fish consumption is the main reason for sub-optimal n-3 LCPUFA intakes of children worldwide.

1.3.2 Dietary n-3 LCPUFA intake in Australian children

1.3.2.1 Dietary recommendations for n-3 LCPUFA intake for Australian children

In Australia, a number of dietary recommendations to consume food sources containing n-3 LCPUFA have been established as part of healthy eating guidelines, as well as for maintaining cardiovascular health. The Australian Dietary Guidelines 2005 recommended that children consume a half to one serving of fish each day, with a serve size equivalent to 80-120 grams of cooked fish fillet, for optimal brain development and cardiovascular health143. The latest revision of the Australian Dietary Guidelines (2013)144, that was based on a comprehensive systematic technical review of the literature provides more specific recommendations for fish consumption, whereby a standard serve size equivalent (serve sizes = 500-600 kJ) for fish is 100 g cooked fish fillet (about 115 g raw weight) or 1 small can of fish (no added salt, not in brine). Recommendation for fish is part of a large recommendation for protein consumption (fish, lean meats, poultry, eggs, tofu, nuts and seeds, and legumes/beans). Recommended minimum number of this food group per day are 1, 1.5 and 2.5 serves for children aged 2-3 y, 4-8 y and 9-18 y, respectively144. Specific quantities depend on age and sex145.

Chapter 1. Introduction

On the other hand, other agencies such as the National Heart Foundation of Australia (NHFA)146,147 have recommendation specifically for fish recognizing the related CVD health benefits. These recommendation are higher than NHMRC (2 to 3 serves of 150 g per serve size of fatty fish per week), as well as food and drinks enriched with marine n-3 or fish capsules for optimal cardiovascular health and prevention of stroke for adults. Children are recommended to follow the adult’s recommendations147.

The National Health and Medical Research Council (NHMRC) of Australia has nutrient reference values like an adequate intake (AI) and a suggested dietary target (SDT) for n-3 LCPUFA. The AI is defined as “The average daily nutrient intake level based on observed or experimentally-determined approximations or estimates of nutrient intake by a group (or groups) of apparently healthy people that are assumed to be adequate” and the SDT is defined as “A daily average intake from food and beverages for certain nutrients that may help in prevention of chronic disease”148. The nutrient reference value for n-3 LCPUFA for children less than 14 years are based on adequate intakes (AI) (Table 1), while for children aged 14-16 years, a suggested dietary target (SDT) has been set at 610 and 430 mg/d for boys and girls, respectively, based on the observed 90th percentile of the population intake148. Hence, the AI reflects the median intakes of the population and is not a recommended intake per se, while SDTs are target intakes for optimal health, rather than prevention of deficiency states. Meyer & Kolanu136 have extrapolated SDTs for n-3 LCPUFA for Australian children younger than 14 years from adjusted energy intakes, by sex and age group (Table 1.3).

A number of ways have been proposed to achieve these targets including nutrition education and behavior modification strategies to increase the consumption of fish and seafood, preferably oily fish, along with n-3 LCPUFA supplementation, or incorporation of foods that are enriched with n-3 LCPUFA, such as certain commercial brands of bread, milk, yoghurt or eggs. Such n-3 LCPUFA enriched bread, egg, milk, yoghurt are available in groceries and supermarkets in Australia. Behaviour modification strategies are required to encourage non-fish eaters to include fish and seafood in their habitual diets, and to overcome reported barriers to intake, for instance undesirable physical properties (smell and bones), difficulties with preparation and cooking, unaffordability, presence of food allergies, and a perceived risk of

Chapter 1. Introduction

pollutants141,149-151. Communication of health messages that encourage children to include more fish in their diet152,153 may be one strategy, but educating parents and caregivers about practical, easy ways to prepare fish dishes may be of higher importance.

Table 1.3: The nutrient reference values for n-3 LCPUFA for Australian children136

Age Sex AI SDT Boys 40* 325# 2-3 y Girls 40* 300# # Boys 55* 400 4-8 y # Girls 55* 350 Boys 70* 510# 9-13 y # Girls 70* 410 Boys 125* 610* 14-16 y Girls 85* 430*

*AI and SDT values obtained from ‘Nutrient Reference Values for Australia and New Zealand148, # These SDT values have been adjusted for total energy intake by age and sex136

1.3.2.2 Situational analysis: dietary n-3 LCPUFA intake in Australian children

Understanding the n-3 LCPUFA intake in the population of children is necessary in order to support optimal health. In-depth nutrient and food analysis can inform the development of appropriate nutrition-related strategies to address poor food habits that exist in the community. Such analysis can also be used to monitor adherence to dietary recommendations as well as to design food policy such as nutrition interventions.

Fish and foods enriched with n-3 are consumed in small quantities in Australian children’s diets136. Reports of Australian national surveys of children conducted in 1985 and 1995 indicated a low intake of fish, with a mean of only 6 to10 g/d154 and 12 to 19 g/d155, respectively. Further, the most recent national nutrition survey of children (Children’s survey 2007) demonstrated approximately 20 % of children consumed fish/seafood on one or both days of the survey (mean = 13 g/d) and only about 1 % of children reported taking fish oil regularly136. To date, there is no information available

Chapter 1. Introduction

regarding intakes of individual n-3 LCPUFA, as well as the respective food sources of these fatty acids from the data collected in the 2007 Children’s survey. This is important because DHA is vital for neurological development57.

Numerous barriers have been reported that may explain the low fish intake in the Australian population156-159. However, these studies contain little information about children’s food preferences.

Inappropriate and inaccurate dietary instruments to assess fish intake141,142,160 may have incorrectly quantified fish consumption patterns among Australian children. The use of the 24-h recall in the previous survey may introduce inherent errors in estimating habitual intake of a food that is consumed episodically, such as fish (will be discussed further in section 1.6). A PUFA FFQ for Australian adults has been previously developed, and showed a valid and reproducible method to estimate n-3 LCPUFA compared to 3d weighed food records161-163 and blood biomarkers161,163. However, a specific dietary instrument is not currently available to assess n-3 LCPUFA in Australian children.

Australia is the leading market for n-3 enriched products, for example bread enriched with n-3164, and Australian consumers claim to have good awareness of the health benefits of n-3 enriched foods165. However, only less than 7 % of Australian children consume foods that are enriched with n-3136. One limitation to recommending an increase in foods enriched with n-3 LCPUFA is the need for the food items to be consumed in large enough quantities to meet the recommendation of 500 mg/d166. This may not be feasible based on individual food preferences and socio-economic or cultural factors, and may be contrary to the recommendation to eat a wide variety of nutritious foods143,144.

1.4 Factors that influence fish and/or seafood consumption

Fish and/or seafood consumption is influenced by many factors which may have important roles in improvement of n-3 LCPUFA intake. Various barriers have been reported in relation to the consumption of fish and seafood in some countries. A study in UK reported barriers for fish consumption including difficulty in buying, preparing and cooking of fish; the belief that fish is expensive; and unpleasant physical properties

Chapter 1. Introduction

of some varieties of fish (e.g. bones and smell)149. A study in Belgium demonstrated that fish bones and price were main barriers, while safety and smell of fish were identified as potential barriers to fish consumption151. A study in Norway has reported that barriers for fish consumption included lack of a supply of fresh supplies, as well as variation in quality, family member dislikes, too few product choices and a dislike of the taste150.

A study in Swedish children demonstrated predictors of intentions to consume fish (e.g. attitudes towards fish, friends’ behaviour and perceived control) and their barriers (e.g. a negative attitude towards both smell and the accompaniments, and fear to finding bones)167. Food neophobia, which is described as ‘avoidance of and reluctance to taste of unfamiliar food’168 is also a predictor of low frequency of fish consumption in Finnish young adults169. Sichert-Hellert141 reported that the barriers to fish consumption in Germany children and adolescents seem to be related to environmental issues (such as the pollutants, dioxins and heavy metals), a large influence of olfactory senses (e.g. taste, smell), the lack of provision either at home or through the school catering service, presence of younger children in the family and a low awareness to the need for an adequate intake of food containing n-3 fats.

It has been reported that health awareness influences fish consumption via three intentions: 1) increasing positive attitude towards fish consumption, 2) higher social pressure from peers or one’s own moral responsibility, and 3) higher conviction of one’s personal ability to buy and prepare fish. These factors have been found to strongly influence intention to consume fish141.

1.5 Foods enriched with n-3 LCPUFA – an alternative source of n-3 LCPUFA intake and their potential health benefits

In countries with traditionally low fish consumption, n-3 enriched foods could play an important role in meeting the recommendations for n-3 LCPUFA for optimal health166. These foods can also be used as sources of n-3 LCPUFA for certain individuals or communities who may not be suitable or convenient to eat common food sources of n-3 LCPUFA (e.g. fish, egg and meat) for whatever reason such as food allergies and

Chapter 1. Introduction

certain religions/cultures as well as vegetarian and vegan. Ovolacto vegetarian may obtain a limited amount of n-3 LCPUFA from eggs, milk and dairy products, while vegan must rely on in vivo biosynthesis of these fatty acids from ALA which it depend on the ratio of ALA and LA from the diet170. An international recommendation of consumer’s need of a daily 500 mg of n-3 LCPUFA for each person have been calculated and showed that current global fish harvest will not supply this recommendation171. Therefore alternative sources are required to meet the recommendation for n-3 LCPUFA for optimal health.

One method to increase n-3 LCPUFA is to incorporate the main sources of n-3 LCPUFA into regularly consumed foods. These include enrichment foods with fish oils, single cell micro algal oils and/or biomass, and under-utilized marine sources e.g. krill15. A number of n-3 enriched foods with the alternative source of n-3 and fish oil e.g. breads, eggs, milk and yoghurt can be found in marketplace in Australia (Rahmawaty & Kolanu, 2009 – unpublished data) and in many countries such as Europe, Canada and South America18.

It has been reported that regular consumption of a variety of n-3 fortified foods (8 meals/day) providing 50 and 150 mg EPA plus DHA per serving increases the daily n-3 LCPUFA intake of Australian adults from 100-200 mg/d to 1000 mg/d172, double the recommended n-3 LCPUFA intakes147. Furthermore, an increase in n-3 LCPUFA concentration in erythrocytes by 35 and 53 % at 3 and 6 months respectively, after supplementation, which is equivalent to an n-3 index of increased from 4 to 7 % over 6 months study172, placing these subjects in lower risk for cardiac death117. These incensements have also been reported to be associated with CVD risks reduction and positively associated with arterial compliance assessment and negatively associated with serum CRP and urinary 11-dehydro-TXB2 excretion172.

Food matrix or carrier food can influence bioavailability and possible physiological effects of n-3 LCPUFA in the systemic circulation or the target organ173. A mini review of currently available published studies on the influence of consumption of products enriched with n-3 LCPUFA174 demonstrated a comparable bioavailability between foods enriched with n-3 and fish oil supplements.

Chapter 1. Introduction

The following section explains the bioavailability and health effects of food products enriched with n-3 LCPUFA that are available in Australia (e.g. bread, egg, milk and yoghurt) as well as other foods that are potentially enriched with n-3 LCPUFA.

Bread enriched with n-3

Bread is a staple for large numbers of people, especially in Western countries such as in Australia. Enrichment of bread with n-3 LCPUFA may have a critically important role because the typical Western diet, which is relatively high in n-6 and low in n-311. Bread is an ideal medium for n-3 because the CO2 produced during dough fermentation protects the oil from oxidation, especially while it is exposed to high temperatures during baking175. Furthermore, baking did not give rise to oxidation of polymerization of fatty acids176.

Liu et al. (2001)177 reported an increase in plasma levels of EPA, DHA and total n-3 FA has been reported of hyperlipidaemia subjects after 2 and 4 weeks by consuming 93 g bread per day containing 1.3 g stable fish oil. Furthermore, the authors also reported a decrease in triglycerides and malondialdehyde (MDA), one of several potential marker of oxidative stress or lipid peroxidation178,179, and an increase in HDL- cholesterol in plasma177.

Microencapsulated fish oil incorporated into bread and also biscuits and soup (providing 0.9 g n-3 PUFA intake per d) was equally effective to that from fish oil capsules (three 1-g capsules per day) in increasing platelet EPA and DHA of healthy females, after 4 weeks intervention176. Consumption of a low dose of n-3 LCPUFA supplemented in bread (20 mg of n-3 LCPUFA per slice, derived from microencapsulated tuna oil on average 6 to 8 slices per day for 3 weeks significantly increased concentration of total n-3LCPUFA (EPA, DPA, DHA) from 2.0 to 2.4 % in total plasma and from 3.9 to 4.3 % in PL fraction180.

Microencapsulation of fish oil into food products such as bread has advantages including: 1) microencapsulation can convert fish oils to powder form, thereby increasing the number and type of food into which it can be incorporated; 2) microencapsulation, in minimising oxidation mainly by avoiding contact between the

Chapter 1. Introduction

fish oil and oxygen in the air, has the potential to extend the shelf life of fish oil and also to improve the palatability of fish oil-enriched food products. The palatability of bread was reported to be good and appearing to be the best vehicle for increasing daily EPA and DHA intakes176.

Eggs enriched with n-3 LCPUFA

Eggs can be enriched with n-3 LCPUFA by the addition of fish meal or seed oils (flax or rapeseed) to animal feeds, and this also increasing the n-3 PUFA in chicken meat181,182. Chicken are able to convert ALA to DHA better than human183.

The food enriched with n-3 consumption demonstrates some positively affects in the plasma lipid profile in humans. Consumption of two eggs enriched with n-3 PUFA for 18 days in healthy volunteers significantly increased in HDL-C and decreased in plasma triglyceride, and tends to elevate ratio of HDL-C/TC and HDL-C/LDL-C184. A study in normolipidemic volunteers eating four n-3 PUFA eggs (from flaxseed-fed hens) per day for 2 weeks showed that no significant changes were observed in concentration of total cholesterol, HDL-cholesterol, or plasma triglyceride, and a significant increase in DHA (33 %) and total n-3 PUFA and significant decrease in ratio n-6 to n-3 FA in the blood platelet phospholipids185. Two weeks consumption of one egg enriched with n-3 per day (from hens fed containing 10 % flaxseed) did not significantly increase concentration of plasma total LDL and LDL cholesterol and triglyceride compared with regular egg, and there was a trend of slightly lower concentration of LDL cholesterol186. A study in hypercholesterolemic subjects for 6 weeks showed that consumption of 2 eggs enriched with n-3 per day has no effect in total LDL-C187.

A randomized double-blind, cross-over study of hypercholesterolemic patients has reported that eating two eggs enriched with n-3 from hens fed flaxseed providing 217 mg of DHA and 629 mg total n-3 LCPUFA per day, significantly increased in EPA plus DHA levels in serum phospholipid by 23 %, and significance correlated to a reduction of fatal ischemic heart diseases risk and did not change levels of lipid and total serum cholesterol188. A double-blind, cross-over study in healthy individual with an age over 45 years concluded that consumption of one egg enriched with n-3 (from hens fed rapeseed oil, and the amount of ALA, EPA and DHA are 9.3, 0.2 and 1.5 % in one egg

Chapter 1. Introduction

respectively) increased ApoA1 and decreased ApoB/ApoA1 ratio, and had no negative impact on blood lipids and inflammatory markers189. Apo-A1 is a major component of HDL that is responsible for carrying cholesterol from arteries, whilst Apo-B is responsible for carrying cholesterol to tissues, and they are accepted as superior markers of CVD risk190. It has also been reported that a high Apo-B/Apo-A1 ratio was associated with metabolic syndrome in obese children191.

An increase in serum lutein, an antioxidant that protects the macula from light- initiated oxidative damage192 has been found in healthy adults193 and in healthy lacto- ovo-vegetarian (LOV) adults194 following intervention by consumption of n-3 enriched eggs. Lutein is highly concentrated in the human eye, in the macula, retina fluid, rod outer segment membranes and the lens of the eye. Together with DHA, lutein can be used for preventing the risk of age-related macular degeneration195.

Milk enriched with n-3 LCPUFA

Milk is a very efficient carrier for fat absorption, due to high dispersion of fat milk in micelles facilitating in a large surface for absorption196. Bioavailability of microencapsulated fish oil incorporated into milkshake197 has been shown similar to that from fish oil capsules after 4-week intervention. Semi-skimmed milk containing n-3 fatty acids from fish oil supplying 0.3 g of DHA plus EPA per day increase in plasma concentrations of DHA plus EPA (30 % on average) together with a significant 19 % reduction in fasting plasma triacylglycerols levels after 6 weeks intervention196. Consumption of 500 mL of the n-3 enriched milk providing DHA and EPA 0.13 g and 0.2 g respectively significantly increased plasma levels of DHA and EPA (on average 30 %) and decreased in plasma levels of homocysteine (13 %), after 2 months intervention198. In another study, an average inclusion of 300 mg EPA plus DHA in milks produced a 25-50 % improvement in plasma levels of fatty acids after a minimum period of 6 weeks114.

Blood lipid effects of the n-3 enriched milks have been reported in healthy children and adults, and also in CVD patients114. All the studies concluded that consumption of milk substituted with EPA plus DHA, PUFA and oleic acid for more than 6 weeks resulted in sustainable reductions blood lipid concentration including total cholesterol (range 4-11 %) and LDL-cholesterol (6-20 %), particularly when baseline

Chapter 1. Introduction

values before intervention were increased114. Another study by Romeo et al.123 reported that consumption of milk enriched with fish oil EPA plus DHA for 5 months significantly decreased endothelial cell activation indices namely ICAM-1 in healthy children aged 8-14 y. Soluble ICAM-1 is associated with future risk factors for atherosclerosis in children199, while high concentration of serum E-selectin is correlated with the development of atherosclerosis and increase in CVD risk in adults200. This suggests that consumption of food enriched with n-3 may reduce CVD risks.

Other study reported that consumption of 200 mL milk fortified with 2 g fish oil twice a week for six months (provided 200 mg EPA plus 1 g DHA each day) significantly increased n-3 LCPUFA in plasma phosphatidylcholine and decreased episode and duration of illness, particularly upper respiratory tract infection such as rhinitis, cold and influenza in Thai school children (9-12 y)201.

Yoghurt enriched with n-3 LCPUFA

Yoghurt enriched with n-3 LCPUFA is possibly a better way to deliver compared to milk. Compared to milk enriched with fish-oil, yoghurt enriched with fish-oil demonstrated less susceptibility to oxidation202, thereby reducing fishy odours and flavour203. Antioxidant peptides released during the fermentation of milk by lactic acid bacteria and/or the lower oxygen content of yoghurt might responsible to the higher oxidative stability of yoghurt204. Technique in adding fish oil for example net fish oil emulsion seemed to increase the stability of yoghurt than those with fish-oil-in-water emulsion203.

A bioavailability study in humans demonstrated that yoghurt drink was the best matrix compared to fitness bar, butter and bread containing fish oil for providing fast absorption of EPA and DHA205 which might be due to the preformed emulsions.

Other foods enriched with n-3 LCPUFA

A number of food products such as dip and juice have also been enriched with n-3 LCPUA and showed positive effect to lipid profile. Consumption n-3 PUFA-enriched dip 100 g/d supplying 1.3-1.4 g of n-3 PUFA (sum of ALA, EPA, DPA and DHA) for 6 weeks increased EPA, DPA and DHA concentrations in plasma lipids by 117, 15 and 80 % respectively, significantly reduced plasma triglyceride level (1.93 mmol/L to 1.27 23

Chapter 1. Introduction

mmol/L) and increased HDL-cholesterol in people with type 2 diabetes mellitus206. A study in healthy children age 4-12 y in US reported an increase in phospholipids DHA contents by 40-50 and 65-70 % after 6 weeks supplementation of 180 mL juice/d fortified with 50 and 100 mg/d microencapsulated-algal DHA respectively113. Algal oil provide an equivalent amount of DHA to bloodstream as cooked salmon, whereas bioavailability of 600 mg/d DHA from salmon or algal oil are equivalent for incorporation into plasma erythrocytes and phospholipids207.

This evidence suggest that a low dose of n-3 LCPUFA can be used as a long-term strategy to achieve the health benefits of n-3 LCPUFA for optimal health.

1.6 Methods to improve dietary intakes in children

Health-promoting behaviour like healthy eating (e.g. n-3 LCPUFA intake) require a thorough understanding of the problem in relation to current gaps and needs to identify determinants of behaviour and environmental conditions208. This process is the first step in health promotion planning which should be (preferably) evidence based to improve the effectiveness of the promoted health behaviour208,209. Therefore, health educators or practitioners should carefully analyse of what is responsible for environmental change, what will determine that change, and how methods and strategies are expected to influence these determinants210.

An evidence-based systemic review of children’s healthy eating interventions has been done in Australia addressed to deliver better health promotion strategies, methods and activities in order to maximise health outcomes211. In this review, a number of important points have been recommended for practitioners, especially in designing nutrition interventions that are likely to be effective to promote healthy eating changes in children. These include focus on: 1) approaching both children and parents in setting they live in, 2) eating, including dietary patterns as well as the taste and enjoyment of food, 3) making small sustainable changes to reach large number of children, 4) development and policy control, and 5) evaluation activities211.

Chapter 1. Introduction

1.6.1 Intervention Mapping – a stepwise approach for designing behaviour changes development

Intervention Mapping (IM) is a systemic evidence-based approach for development behaviour changes210,212,213. It provides health education planner with a framework for effective decision making at each step in the intervention development process by integrating theory, empirical findings from the literature, and information collected from the target population210. IM argues that for effectiveness of health promotion intervention, the first important process is to translate the health related behaviour (e.g. high fat intake) into a health-promoting behaviour (e.g. fat reduction) and then search the determinants of the required change208. The ability to influence mediator of nutrition behaviour change, for example environmental factors (e.g. food availability and accessibility) would determine the successful of the targeted behaviour change214. An IM approach is characterised by a sequence planning process including: (1) analysis of the problem and specifying mediators in the environment and their role on the target behaviour or environmental change; (2) integrating specific theories with specific determinants changing; (3) providing behaviour changes in a system; (4) evaluation outcome in relation to program objectives. Applying Intervention Mapping may improve interventions by providing more detail and guidance for the planning process and the logic of change215.

The IM method has been used for over two decades, and proved to be a useful tool for designing healthy behaviour changes210, such as in designing and improving fruit and vegetable intakes for school children216,217. A community intervention study has been conducted in Western Australia addressed to encourage regular seafood consumption218. The study used participatory approach which is an application of IM219 to develop, implement and evaluate of coordinated and specific educational training resources across a number of sector, including school curriculum from primary through to vocational sectors, point of sale resources for consumers, and condition-specific resources and information for health professionals. The results of the study demonstrated an increase in seafood consumption (as measured by sales) of 24 % over the intervention period and a residual increase 15% in the proceeding month218.

Chapter 1. Introduction

1.6.2 Parent’s influence to children’ eating habit

Food consumption of children is stimulated by food preferences220. These are influenced by a number of factors; one such is parent’s preferences221. Parents are considered to have a great influence on development of children’s eating habits, where they not only create a condition for children’s beginning familiarity with food and eating, but they also affect their children’s food consumption through children’s imitating their parents’ eating pattern, appetite and food selection222. Parents can also provide an initial model for children in terms of how to promote healthier eating habit223. Children’s fish consumption patterns are similar to that of their mother224. Home environment that involve parental feeding practice influence the development of food-acceptance patterns of a child that affects food preferences later in life225. Involvement of adolescents (9-14 years old) in family dinner impacted quality of diet toward healthy eating behaviour226. Eighty-four per cent of parents feel that they have most responsibility for ensuring that their children eat a healthy diet, only 7 % believe that the government has the most responsibility, and less than 3 % believe that food retailers have the most responsibility227. Therefore, it is suggested that interventions targeting healthy eating behaviours in children should involve family members or modify home environments, whereas parents and other family members in the household also consuming the healthy foods222,228, including fish and food enriched with n-3 LCPUFA.

1.6.3 Dietary modelling as a strategy to meet dietary recommendation

Food enrichment with specific nutrients such as n-3 enriched food can lead to relatively rapid changes in the specific nutritional status of a community, and is a cost-effective public health intervention. However the items need to be consumed in adequate amount by a large proportion of target individuals in the population and the levels of fortification must be high enough to substantially increase the intakes166,229.

Three aspects have been proposed which may explain the low level of compliance on dietary recommendations230. These include consumer confusion with ongoing delivery of sometimes conflicting dietary information, advice that is not adequately individualized to address personal health concerns and failure to integrate experimental evidence with real-life situations230. Consumer-based qualitative studies have concluded

Chapter 1. Introduction

that consumers want realistic and personalized guideline to help them adopt more healthful eating habits. Hence, it is suggested to provide concrete examples and explicit directions for supporting progressive changes in food selection, one such using dietary modelling231.

A dietary modelling is a conceptual or technical approach which translates specific dietary recommendations or nutrient reference values into dietary models. It describes the amounts of various foods needed to meet the estimated nutrient requirements of groups of people145. It could also be used to appraise the effectiveness of a dietary recommendation to increase specific dietary intake in the community.

A food pattern modelling study has been used in America to develop a dietary guideline for Americans, and one of the results is fish consumption modelling232. The study demonstrated that consumption of approximately 227 gram of all type of fish or fish high in n-3 per week results in average intake approximately 31 gram of fish per day, almost triple current fish consumption, or more than 10 times current fish high in n-3 consumption among Americans. Beside, 2 servings of all type of fish per week would provide approximately 200 mg/d of n-3 LCPUFA (EPA + DHA), while 2 servings of fish high in n-3 would contribute 500 mg/d of n-3 LCPUFA. It has been reported that substitution of canola oil-based margarine and spreads in dietary modelling increased compliance with the dietary recommendations for fatty acid231. In Australia, the latest document of dietary modelling is available on the Modelling System to inform the Australian guide to healthy eating145.

1.6.4 Food pattern analysis

In general, individuals consume meals or dishes consisting of a variety of foods with complex combinations of nutrients that are likely to be interactive or synergistic, therefore it is necessary to examine the effect of overall diet instead a single food. Studies of individual foods and/or nutrients have often failed to provide conclusive evidence of causality, primarily due to the interactions that exist between them, which cannot be readily accounted for in any statistical analysis. Exploiting dietary pattern methodologies to the full could help to unravel the complex relationships between the foods we ingest and the myriad of health outcomes experienced.

Chapter 1. Introduction

Recently, the analysis of pattern of food choices (sometimes called food pattern analysis) has an increase attention of researchers because it focuses on whole diet which makes it easier to examine the interaction between foods and nutrients in relation to health233. Food pattern analysis also can be used to evaluate dietary guidelines or dietary recommendations due to the capability of the food pattern analysis to generate a boarder picture of food consumption in a population234. It may also be useful in designing community programs in relation to specific health problem. The food factor analysis has been widely used to examine the relationship of the overall dietary patterns with the risk of chronic diseases235,236 and nutritional status233,237.

Food patterns are typically examined by statistical methods for example factor analysis. Principle-component analysis (PCA) is commonly used to define dietary patterns using food consumption information to identify common underlying dimensions (factors or patterns) of food intake238,239. PCA extracts underlying patterns that explain the variation in how people eat where the eating pattern of a person is represented by combination of all patterns (through component score) in PCA238. This method aggregates specific food items based on the degree to which these food items are correlated with each another239. A summary score for each pattern is then derived and can be used to examine relationships between various eating patterns and outcomes of interest, such as coronary heart disease and other chronic diseases234,239. A study by Loiret et al. suggested the important of distinguishing dietary patterns between sexes, and not just in scores calculated from the PCA but the patterns themselves and the foods that are associated with them240. This is particularly important when looking within a family setting and more specifically when examining the influences of parental lifestyle on the health and well-being of their children240.

1.7 Methodological issues associated with assessing fish and n-3 LCPUFA intake in children

An appropriate dietary instrument to assess n-3 LCPUFA intake is needed to provide accurate estimation of n-3 LCPUFA intakes. Since fish is not regularly consumed by most people, assessment n-3 LCPUFA intake should consider to the frequency of fish consumption over the past specific period of time in the community, instead using a single day report. If the report happens on a non-consumption day of fish or seafood, many fish or seafood consumers will report zero intakes on the day recall. The food 28

Chapter 1. Introduction

frequency questionnaire might provide valuable information, especially to improve estimates of n-3 LCPUFA intakes as well as could be a useful covariate in estimating the probability of consumption and amount consumed241.

1.7.1 Food frequency questionnaire as a dietary assessment method for foods periodically consumed

FFQ is defined as “a questionnaire in which the respondent is presented with a list of foods and is required to say how often each is eaten in broad terms such as x times per day/per week/per month, etc. Foods list are usually chosen for the specific purposes of a study and may not assess total diet”242. The food frequency questionnaire consists of a list of foods and a selection of options relating to the frequency of consumption of each of the foods listed (e.g. times per day, daily, weekly, monthly). FFQs normally ask about intake within a given time frame (e.g. in the past 2-3 months, 1 year or longer) and therefore aim to capture habitual intake. Although there are difficulties implicit in calculating the absolute nutrient intake of individuals from food frequency questionnaires, they are useful for gathering information on groups of individuals as well as for looking at habitual intake of a range of foods.

1.7.1.1 Pros and cons of FFQ compared to other dietary instruments

Pros of FFQ

Compared to other dietary instruments such as 24 hour recalls and dietary history questionnaire, FFQ has a number of advantages. FFQ are designed to collect dietary information from large numbers of individuals (100 individuals or more) and are normally self-administered, though interviewer administered and telephone interview are possible modifications243, and therefore suitable for large scale surveys and can be posted. Specific time frame given in the FFQ can capture habitual intake and give low respondent burden with require a minimal amount of time from the respondent.

FFQ has a number of superiors to analyse food occasionally consumed such as fish/seafood, because it involves a selection of specific food sources (foods list) and options relating to the frequency of consumption of each of the foods listed over a specified time period. The use of the 24-h dietary recall in estimating habitual fish or seafood consumption may introduce inherent errors, leading to potential reporting bias. 29

Chapter 1. Introduction

Cons of FFQ

Food frequency questionnaire, as well as other dietary measurement methods such as 24-hour food recalls and dietary history questionnaire involves various aspects of individual cognition, especially in remembering, imagining, and recognizing and interpreting the type and amount of food consumed.

The number of food choices on the FFQ may lead to overestimate consumption of respondent244. A number studies in the population of children reported that FFQ shows overestimate absolute intake compared to other dietary assessments such as two 24-h food recall245,246, 2d food diary247 and 3d food record248. Four day weighed food record by children aged 8- 11 years showed approximately equal to FFQ compared to parents’ report249.

FFQ is very sensitive to cultural and dietary practice250 and only suitable for one population and for intake estimation of certain nutrients and might be invalid when applied to another. Therefore, it is crucial to validate the FFQ in relation to reference method in a specific population to provide accurate and actual intakes data.

1.7.1.2 Validation of FFQ

It is accepted that measurement of dietary intakes, mostly depends on individual cognitive aspects, and the intakes values obtained must be considered as values of a latent variable, which absolutely reflects a fact, but imposible without errors251. Thus, it is crucial to validate the dietary instruments to provide accurate and actual dietary intakes data.

A validation study or called a relative validation/calibration study is used to compare a dietary method with another dietary method to evaluate whether what they actually measured truly reflect what they are supposed to measure252. There are two major points that must be answered in the validation study including the strength of association of the measurement with the true variable of interest and “do measurements reflect to nothing else but the variable of interest (question of specification)251. A validation study can be done by comparing the two methods of dietary measurements in which one is considered as gold standard e.g. FFQ is compared against multiple days of diet recalls or records. It also can be done by adding a biomarker measurement 30

Chapter 1. Introduction

considered as an objective measurement in the first method. The validation study by comparing two dietary methods has a limitation because the two dietary instruments used in the first method to have errors that are related each other because they have the same source of random error253. Additionally, the random error of the two dietary methods is not independent but correlate with each other, which may lead to an overestimation or underestimation for the validity of the FFQ254.

The method of validation by adding a biomarker measurement has been broadly used in the validation study, particularly for development of FFQ to assess n-3 LCPUFA intakes in adults population163,255-259. However, there have been limited studies conducted in children. A study in USA for validating an FFQ using parental report for children aged 1-11 years using erythrocyte membrane as a biomarker of n-3 LCPUFA258. The study demonstrated that intakes of total n-3 fatty acids (beta=0.52, p<0.0001, rho=0.23) and marine PUFAs (beta=1.62, p<0.0001, rho=0.42), as a percent of total fat in the diet, were associated with percent of n-3 and marine PUFAs in the erythrocyte membrane258. The use of biomarkers of fatty acid status also has limitations when using it for the validation of an FFQ, as biomarkers do not only reflect the fatty acid intake but also the endogenous synthesis.

1.7.2 Development of an FFQ for children

Children’s self reporting of dietary intakes are complex tasks, involving individual cognitive aspects in relation to perception, conception and memory260. Hence, the designing of questionnaire for children was suggested age-sensitive and generally combined with numerous strategies to improve their memory261.

It is reported that children’s estimates of portion size using age-appropriate food photographs were significantly more accurate (an underestimation of 1 % on average) than estimates using photograph designed for use with adult (an overestimates of 45 % on average)262. Moreover, the use of photographs over three dimensional models is systematically more accurate to estimate children’s serving size of age 4-16 years263. Practice (45 minutes) about estimation portion size in children of 9-10 years, significantly improved their ability to quantify food264.

Chapter 1. Introduction

Estimation of portion size is a necessary factor in self-report of dietary intake. According to Nelson et al., there are 3 elements affect portion size reports, which are perception, conceptualization and memory. The perception is defined as the ability to relate a food amount present in reality to the amount presented by a portion-size aid; conceptualization is the ability to develop a mental picture of a food portion not actually present and relate it to a portion size-aid; and memory is the ability to accurately recall an amount of food eaten and can affect conceptualization260.

1.8 Summary of evidence and gaps in the literature

Evidence exists for the health benefits of n-3 LCPUFA to support the normal growth and cognitive development of children and promising CVD risks prevention later in life. The n-3 LCPUFA must be obtained from pre-formed n-3 LCPUFA (not from the conversion from ALA to DHA) which are provided in fish and food enriched with n-3 LCPUFA. However, numerous barriers to consume fish lead to a failure to meet the reference values for fish consumption among the children population of different countries as well as Australia.

Home environment like parent’s attitude in preparing food for their family may affect food preference of their children, including fish consumption. Therefore, interventions targeting healthy eating behaviours in children should modify home environments such as by an approach to parents. Modifying factors that prevent and encourage parents to serve fish and n-3 LCPUFA enriched foods for their family may improve n-3 LCPUFA intake in children.

In countries with traditionally low fish consumption such as Australia, foods enriched with n-3 LCPUFA may play an important role to meet the dietary recommendations for n-3 LCPUFA. Concrete examples and explicit directions for supporting progressive changes in selection of appropriate foods containing n-3 LCPUFA is needed to help consumer in meeting dietary recommendation for n-3 LCPUFA.

Food pattern analysis is useful to examine interaction between foods and nutrients in a whole diet instead a single food and it is recommended to evaluate recommended dietary intake in a population in relation to health. The capability of this method to

Chapter 1. Introduction

capture the boarder picture of dietary patterns in a community can be used to support designing nutrition policies or intervention programs in order to promote healthy eating changes.

A food frequency questionnaire is suitable to assess n-3 intake in an adult population and may be applied in population of children.

This review has identified the following gaps in the literature, which will be addressed in this thesis:

- To date, there is a limited data available for EPA, DPA, and DHA intake as well as their respective food sources in Australian children. - There is a little research measuring the barriers and the promoting factors to the consumption of fish and foods enriched with n-3 LCPUFA in the Australian family with young children (primary school age children). - There is a lack of research examining the effectiveness of the recommendation to consume foods enriched with n-3 LCPUFA and comparing that to recommended intake for n-3 LCPUFA for Australian children. - There is no research on the relation of food patterns with n-3 LCPUFA intake in Australian children. - There is no specific validated FFQ that can assess dietary n-3 LCPUFA intakes in Australian children.

1.9 Aims and hypothesis

The central aims of this thesis are: 1) to assess current EPA, DPA and DHA intakes and mediators to improve the intakes of these fatty acids and 2) to develop better tools for monitoring dietary intake in Australian children’s diet.

The specific aims of this thesis are: - Study 1: to assess dietary intakes of EPA, DPA, and DHA and the contribution of food sources of these fatty acids in the Australian children’s diet by sex, age groups, and fish-eater status. - Study 2: to identify behaviour and factors promoting and preventing fish and seafood consumption.

Chapter 1. Introduction

- Study 3: to design dietary modelling which would provide choices for the consumers to improve their n-3 LCPUFA intakes. - Study 4: to examine food patterns in relation to n-3 LCPUFA intake in Australian children. - Study 5: to design and asses the validity of an FFQ aimed to assess n-3 intake of Australian children.

The central hypothesis of this thesis is in depth nutrient analysis of n-3 LCPUFA intake improve methods for monitoring the intakes of n-3 LCPUFA in Australian children. An overview of the studies included in this thesis and the relationship to the central hypothesis is shown in Figure 1.2.

Chapter 1. Introduction

Preliminary Study: Supermarket survey of n-3 LCPUFA enriched foods

Study 1: Study 3: Dietary intake and food sources Dietary modelling of n-3

of EPA, DPA, DHA LCPUFA

- Study 2: Study 4: Consumers’ behaviour to fish Food pattern analysis in and n-3 LCPUFA enriched relation to n-3 LCPUFA foods consumption intakes

Central hypothesis :

- In depth nutrient analysis of n-3 LCPUFA intake

improve methods for monitoring the intakes of

n-3 LCPUFA in Australian children -

-Study 5:

The development of an n-3 - LCPUFA FFQ

: Research flow : Providing feedback

Figure 1.2: Overview of thesis studies in relation to the central hypothesis

1.10 Thesis structure

Chapter 2 provides the result of analyses determining the dietary intake and food sources of EPA, DPA and DHA in Australian children. The - results of these analyses are then used to develop the subsequent steps in this thesis.

Chapter 3 introduces and provides the findings of the fish and food enriched with n-3 LCPUFA survey. This survey investigates barriers and factors promoting fish and n- 3 enriched foods consumption in Australian family with young children. 35

Chapter 1. Introduction

Chapter 4 and 5 introduces and provides the results of analysis investigating the effectiveness of recommendation to consume food enriched with n-3 LCPUFA in order to meet the nutrient reference values for the n-3 LCPUFA (chapter 4) and the food pattern analysis in relation to the n-3 LCPUFA intake (chapter 5).

Chapter 6 introduces and provides the development and validation of an FFQ to assess n-3 LCPUFA in Australian children aged 9 – 13 years.

Chapter 7 is a summary of the findings in the previous chapters in order to provide advice and potential nutrition intervention strategies addressed to improve n-3 LCPUFA intakes in Australian children. Conclusions and recommendation for future research are summarised in this chapter.

References

1 Burr GO, Burr MM (1929) New deficiency disease produced by the rigid exclusion of fat from the diet, J Bio Chem 132, 345-367.

2 Bang, HO, Dyerberg J, Hjorne N (1976) The composition of food consumed by Greenland Eskimos. Acta Med Scand 200, 69-73.

3 Dyerberg J, Bang HO, Stoffersen E, Moncada S, Vane JR. (1978) Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 2, 117- 119.

4 Bang HO, Dyerberg JE (1980) Lipid metabolism and ischaemic heart disease in Greenland Eskimos. Adv Nutr Res 3, 1-22.

5 GISSI-Prevenzione Investigators (1999) Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevention trial. Lancet 354, 447-455.

6 Yokoyama M, Origasa H (2003) JELIS Investigators. Effects of eicosapentaenoic acid on cardiovascular events in Japanese patients with hypercholesterolemia: rationale, design, and baseline characteristics of the Japan EPA Lipid Intervention Study (JELIS). Am Heart J 146, 613-620.

7 Australian Bureau of Statistics (2012) Causes of Death, Australia, 2010. Canberra: Australian Bureau of Statistics. (ABS Cat. No. 3303.0).

8 Venn AJ, Thomson RJ, Schmidt MD, Cleland VJ, Curry BA, Gennat HC et al. (2007) Overwight and obesity from childhood to adulthood: a follow-up of participants in the 1985 Australian School Health and Fitness Survey. Med J Aust 186, 458-460.

Chapter 1. Introduction

9 Cancer Council Australia, Heart Foundation (2011) High school students graduating to be tomorrow’s cancer and cardiovascular patients. Available at: http://www.heartfoundation.org.au/SiteCollectionDocuments/9FEB10%20-%20 MEDIA%20RELEASE%20-%20national%20student%20survey.pdf.

10 Ratnayake WMN, Galli C (2009) Fat and fatty acid terminology, methods of analysis and fat digestion and metabolism: A background review paper. Ann Nutr Metab 55, 8-43.

11 Das UN (2006) Essential fatty acids: biochemistry, physiology and pathology, Biotec J 1, 420-439.

12 Emken EA, Adlof RO, Gulley RM (1994) Dietary linoleic acid influences desaturation and acylation of deuterium-labeled linoleic and linolenic acids in young adult males. Biochim Biophys Acta 1213, 277- 288.

13 Pawlosky RJ, Hibbeln, Novotny JA, Salem N (2001) Physiological compartmental analysis of α-linolenic acid metabolism in adult humans. J Lipid Res 42, 1257-1265.

14 Burdge GC, Finnegan YE, Minihane AM, Williams CM, Wootton SA (2003) Effect of altered dietary n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13 C]α-linolenic acid to longer-chain fatty acids and partitioning towards β-oxidation in older men. Br J Nutr 90, 311-321.

15 Nichols PD, Petrie J, Singh S (2010) Review Long-Chain Omega-3 Oils–An Update on Sustainable Sources. Nutrients 2, 572-585.

16 Howe P, Meyer B, Record, Baghurst K (2006) Dietary intake of long-chain ω-3 polyunsaturated fatty acids: contribution of meat sources. Nutrition 22, 47-53.

17 Meyer BJ, Mann NJ, Lewis JL, Milligan GC, Sinclair AJ, Howe PRC (2003). Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids. Lipids 38, 391-398.

18 Whelan J, Rust C (2006) Innovative dietary sources of n-3 fatty acids. Annu Rev Nutr 26, 75-103.

19 Kitessa SM, Gulati SK, Simos GC, Ashes JR, Scott TW, Fleck E, Wynn PC (2004) Supplementation of grazing dairy cows with rumen-protected tuna oil enriches milk fat with n-3 fatty acid without affecting milk production or sensory characteristics. Br J Nutr 91, 271-277.

20 Wood JD, Enser M, Fisher AV, Nute GR, Richardson RI, Sheard PR (1999) Manipulating meat quality and composition. Proc Nutrition Soc 58, 363-370.

21 Robert SS, Singh SP, Zhou XR, Petrie JR, Blackburn SI, Mansour PM, Nichols PD, Liu Q, Green AG (2005) Metabolic engineering of Arabidopsis to produce nutritionally important DHA in seed oil. Func Plant Biol 32, 473-479.

Chapter 1. Introduction

22 Truksa M, Wu G, Vrinten P, Qiu X (2006) Metabolic engineering of plants to produce very long-chain polyunsaturated fatty acids. Transgenic Res 15, 131-137.

23 Cho HP, Nakamura M, Clarke SD (1999) Cloning, expression, and nutritional regulation of the mammalian D-6 desaturase. J Biol Chem 274, 471-477.

24 Aki T, Shimada Y, Inagaki K, Higashimoto H, Kawamoto S, Shigeta S, Ono K, Suzuki O (1999) Molecular cloning and functional characterization of rat D-6 fatty acid desaturase. Biochem Biophys Res Commun 255, 575-579.

25 Cho HP, Nakamura M, Clarke SD (1999) Cloning, expression, and fatty acid regulation of the human D-5 desaturase. J Biol Chem 274, 37 335-37 339.

26 Leonard AE, Kelder B, Bobik EG, Chuang LT, Parker-Barnes JM, Thurmond JM, Kroeger PE, Kopchick JJ, Huang YS, Mukerji P (2000) cDNA cloning and characterization of human D5-desaturase involved in the biosynthesis of arachidonic acid. Biochem J 347 , 719-724.

27 Vermunt SH, Mensink RP, Simonis MM, Hornstra G (2000) Effects of dietary alpha-linolenic acid on the conversion and oxidation of 13C-alphalinolenic acid. Lipids 35, 137-1 42.

28 Marshall LA, Johnston PV (1982) Modulation of tissue prostaglandin synthesizing capacity by increased ratios of dietary alpha-linolenic acid to linoleic acid. Lipids 17, 905-913.

29 Larsson SC, Kumlin M, Ingelman-Sundberg M, Wolk A (2004) Dietary long-chain n-3 fatty acids for the prevention of cancer: a review of potential mechanisms. Am J Clin Nutr 79, 935-945.

30 Calder PC, Grimble RF (2002) Polyunsaturated fatty acids, inflammation and immunity. Eur J Clin Nutr 56, S14-S19.

31 Kidd PM (2007) Omega-3 DHA and EPA for cognition, behaviour and mood: Clinical findings and structural-functional synergies with cell membrane phospholipids. Altern Med Rev 12, 207-227.

32 Pawlosky RJ, Hibblen JR, Lin Y, Goodson S, Riggs P, Sebring N, Brown GL, Salem N Jr (2003) Effects of beef- and fish-based diets on the kinetics of n-3 fatty acid metabolism in human subjects. Am J Clin Nutr 77, 565-572.

33 Holman RT, Johnson SB, Hatch TF (1982) A case of human linolenic acid deficiency involving neurological abnormalities.AmJ Clin Nutr 35, 617-623.

34 Bjerve K, Mostad IL, Thoresen L (1987) Alpha-linolenic acid deficiency in patients on long-term gastric-tube feeding: estimation of linolenic acid and long-chain unsaturated n-3 fatty acid requirement in man. Am J Clin Nutr 45, 66-77.

Chapter 1. Introduction

35 Phinney SD, Odin RS, Johnson SB, Holman RT (1990) Reduced arachidonate in serum phospholipids and cholesteryl esters associated with vegetarian diets in humans. Am J Clin Nutr 51, 385-392.

36 Rosell MS, Lloyd-Wright Z, Appleby PN, Sanders TA, Allen NE, Key TJ (2005) Long-chain n-3 polyunsaturated fatty acids in plasma in British meat-eating, vegetarian, and vegan men.AmJ Clin Nutr 82, 327-334.

37 Chan JK, McDonald BE, Gerrard JM, Bruce VM, Weaver BJ, Holub BJ (1993) Effect of dietary linolenic acid and its ratio to linoleic acid on platelet and plasma fatty acids and thrombogenesis. Lipids 28, 811-7.

38 Fokkema MR, Brouwer DA, Hasperhoven MB, Martini IA, Muskiet FA (2000) Short-term supplementation of low-dose gamma-linolenic acid (GLA), alpha- linolenic acid (ALA), or GLA plus ALA does not augment LCP omega 3 status of Dutch vegans to an appreciable extent. Prostaglandins Leukot Essent Fatty Acids 63, 287-292.

39 Harper C, Edwards M, DeFilipis A, Jacobson T (2006) Flaxseed oil increases the plasma concentrations of cardioprotective (n-3) fatty acids in humans. J Nutr 2006 136, 83-87.

40 Vlaardingerbroek H, Hornstra G, de Koning TJ, Smeitink JAM, Bakker HD, de Klerk HBC et al. (2006) Essential polyunsaturated fatty acids in plasma and erythrocytes of children with inborn errors of amino acid metabolism. Mol Genet Metab 88, 159-165.

41 Aldámiz-Echevarría L, Bilbao A, Andrade F, Elorz J, Prieto JA, Rodríguez-Soriano J (2008) Fatty acid deficiency profile in children with food allergy managed with elimination diets. Acta Paediatr 97, 1572-1576.

42 Voss A, Reinhart M, Sankarappa S, Sprecher H (1991) The metabolism of 7, 10,13,16,19-docosapentaenoic acid to 4,7,10,13,16,19-docosahexaenoic acid in rat liver is independent of a 4-desaturase. J Bio Chem 266, 19995-20000.

43 Wang Y, Botolin D, Christian B, Busik J, Xu J, Jump DB (2005) Tissue-specific, nutritional and developmental regulation of rat fatty acid elongases. J Lipid Res 46, 706-715.

44 Rosenthal MD, Hill JR (1986) Elongation of arachidonic and eicosapentaenoic acids limits their availability for thrombin-stimulated release from the glycerolipids of vascular endothelial cells. Biochim Biophys Acta 875, 382-391.

45 SanGiovanni JP, Chew EY (2005) The role of omega-3 long-chain polyunsaturated fatty acids in health and disease of the retina. Progress in Retinal and Eye Research 24, 87-138.

Chapter 1. Introduction

46 Li D, Bode O, Drummond H, Sinclair AJ (2002) Omega-3 (n-3) fatty acids. In Gunstone FD, ed. Lipids for functional foods and nutraceuticals, 225-262. The Oily Press, UK.

47 Serhan CN, Hong S, Gronert K, Colgan SP, Devchand PR, Mirick G, Moussignac RL (2002) Resolvins: a family of bioactive products of omega-3 fatty acid transformation circuits initiated by aspirin treatment that counter proinflammation signals. J Exp Med 196, 1025-1037.

48 Bazan NG (2007) Omega-3 fatty acids, pro-inflammatory signaling and europrotection. Curr Opin Clin Nutr Metab Care 10, 136-141.

49 Lee JY, Hwang DH (2008) Dietary fatty acids and eicosanoids. In Chow CK, ed. Fatty Acids in Foods and Their Health Implications, 713-739. CRC Press, London, UK.

50 Serhan CN, Arita M, Hong S, Gotlinger K (2004) Resolvins, docosatrienes, and neuroprotectins, novel omega-3-derived mediators, and their endogenous aspirin- triggered epimers. Lipids 39, 1125-32.

51 Serhan CN, Yang R, Martinod K, Kasuga K, Pillai PS, Porter TF, Oh SF, Spite M (2009) Maresins: novel macrophage mediators with potent antiinflammatory and proresolving actions. J Exp Med 206, 15-23.

52 Madsen L, Petersen RK, Kristiansen K (2005) Regulation of adipocyte differentiation and function by polyunsaturated fatty acids. Biochim Biophys Acta 1740, 266-286.

53 Wallace JMW, McCabe AJ, Robson PJ, Keogh MK, Murray CA, Kelly PM, Marquez-Ruiz G, McGlynn H, Gilmore WS, Strain JJ (2000) Bioavailability of n-3 polyunsaturated fatty acids (PUFA) in food enriched with microencapsulated fish oil. Ann Nutr Metab 44, 157-162.

54 Calder PC (2001) The scientific basis for fish oil supplementation in rheumatoid arthritis. In Food and Nutritional Supplements -Their Role in Health and Disease, ed. Ransley JK, Donnelly JK, Read NW, 175-197. Berlin, Springer.

55 Kaur G, Cameron-Smith D, Garg M, Sinclair AJ (2011) Docosapentaenoic acid (22:5n-3): A review of its biological effects. Prog Lipid Res 50, 28-34.

56 Kanayasu-Toyoda T, Morita I, Murota S (1996) Docosapentaenoic acid (22:5, n-3), an elongation metabolite of eicosapentaenoic acid (20:5, n-3), is a potent stimulator of endothelial cell migration on pretreatment in vitro. Prostag Leukotr Essent Fatty Acids 54, 319-325.

57 McCann JC, Ames BN (2005) Is docosahexaenoic acid (DHA), an n-3 long chain polyunsaturated fatty acid, required for normal brain function? An overview of evidence from cognitive and behavioral tests in humans and animals: a review. Am J Clin Nutr 82, 281-295.

Chapter 1. Introduction

58 Else PL, Hulbert AJ (2003) Membranes as metabolic pacemakers. Clin Exp Pharmacol Physiol 30, 559-564.

59 Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P (2002) Molecular Biology of the Cell. 4th ed. New York, NY, Garland Science.

60 Fliesler SJ, Anderson RE (1983) Chemistry and metabolism of lipids in the vertebrate retina. Prog Lipid Res 22, 79-131.

61 McNamara RK, Carlson SE (2006) Role of omega-3 fatty acids in brain development and function: Potential implications for the pathogenesis and prevention of psychopathology. Prostaglandins Leukot Essen Fatty Acids 74, 329- 49.

62 Giusto NM, Pasquaré SJ, Salvador GA, Castagnet PI, Roque ME, Ilincheta de Boschero MG (2000) Lipid metabolism in vertebrate retinal rod outer segments. Prog Lipid Res 39, 315-391.

63 Clandinin MT, Chappell JE, Leong S, Heim T, Swyer PR, Chance GW (1980) Intrauterine fatty acid accretion rates in human brain: implications for fatty acid requirements. Early Hum Dev 4, 121-129.

64 Martinez M (1992) Tissue levels of polyunsaturated fatty acids during early human development. J Pediatr 120, S129-S138.

65 Lauritzen L, Hansen HS, Jùrgensen MH, Michaelsen KF (2001) The essentiality of long chain n-3 fatty acids in relation to development and function of the brain and retina. Prog Lipid Res 40, 1-94.

66 Carver JD, Benford VJ, Han B, Cantor AB (2001) The relationship between age and the fatty acid composition of cerebral cortex and erythrocytes in human subjects. Brain Res Bull 56, 79-85.

67 Svennerholm L (1968) Distribution and fatty acid composition of phosphoglycerides in normal human brain. J Lipid Res 9, 570-579.

68 Hoffman DR, Birch EE, Birch DG, Uauy R, Castaneda Y, Lapus MG, Wheaton DH (2000) Impact of early dietary intake and blood lipid composition of long-chain polyunsaturated fatty acids on later visual development. J Pediatr Gastroenterol Nutr 31, 540-553.

69 Schuchard JP, Huss M, Stauss-Grabo M, Hahn A (2010) Significance of long- chain polyunsaturated fatty acids for the development and behaviour of children. Eur J Pediatr 169, 149-164.

70 Kim JL, Winkvist A, Åberg MA, Åberg N, Sundberg R, Torén K, Brisman J (2010) Fish consumption and school grades in Swedish adolescents, a study of the large general population. Acta Paediatrica 99, 72-77.

Chapter 1. Introduction

71 Åberg MA, Åberg N, Brisman J, Sundberg R, Winkvist A, Torén K (2009) Fish intake of Swedish male adolescents is a predictor of cognitive performance. Acta Paediatr 98, 555-560.

72 Theodore RF, Thompson JMD, Waldie KE, Wall C, Becroft DM, Robinson E, Wild CJ, Clark PM, Mitchell EA (2009) Dietary patterns and intelligence in early and middle childhood. Intelligence 37, 506-513.

73 Birberg-Thornberg U, Karlsson T, Gustafsson PA, Duchén K (2006) Nutrition and theory of mind-The role of polyunsaturated fatty acids (PUFA) in the development of theory of mind. Pros Leuko Ess Fatty Acids 75, 33-41.

74 Ryan AS, Nelson EB (2008) Assessing the effect of docosahexaenoic acid on cognitive functions in healthy, preschool children: a randomized, placebo- controlled, double-blind study. Clin Pediatr 47, 355-362.

75 Dalton A, Wolmarans P, Witthuhn RC, Stuijvenberg ME, Swanevelder SA, Smuts CM (2009) A randomised control trial in schoolchildren showed improvement in cognitive function after consuming a bread spread, containing fish flour from a marine source. Prostaglandins Leukot Essent Fatty Acids 80, 143-149.

76 The NEMO Study Group (2007) Effect of a 12-mo micronutrient intervention on learning and memory in well-nourished and marginally nourished school-aged children: 2 parallel, randomized, placebo-controlled studies in Australia and Indonesia, Am J Clin Nutr 86, 1082-1093.

77 Muthayya S, Eilander A, Transler C, Thomas T, Knaap H, Srinivasan K, Klinken BJW, Osendarp SJM, Kurpad AV (2009) Effect of fortification with multiple micronutrients and n23 fatty acids on growth and cognitive performance in Indian schoolchildren: the CHAMPION (Children’s Health and Mental Performance Influenced by Optimal Nutrition) Study. Am J Clin Nutr 89, 1766-1775.

78 Neuringer M, Connor WE, Lin DS, Barstad L, Luck S (1986) Biochemical and functional effects of prenatal and postnatal omega 3 fatty acid deficiency on retina and brain in rhesus monkeys. Proc Natl Acad Sci USA 83, 4021-4025.

79 Reisbick S, Neuringer M, Hasnain R R, Connor WE (1994) Home cage behavior of rhesus monkeys with long-term deficiency of omega-3 fatty acids. Physiol Behav 55, 231-239.

80 Crawford MA (1993) The role of essential fatty acids in neural development: implications for perinatal nutrition. Am J Clin Nutr 57 (Suppl), 703S-709S.

81 Ahmad SO, Park JH, Radel JD, Levant B (2008) Reduced numbers of dopamine neurons in the substantia nigra pars compacta and ventral tegmental area of rats fed an n-3 polyunsaturated fatty acid-deficient diet: a stereological study. Neurosci Lett 38, 303-307

Chapter 1. Introduction

82 McNamara RK, Able J, Liu Y, Jandacek R, Rider T, Tso P, Lipton JW (2009) Omega-3 fatty acid deficiency during perinatal development increases serotonin turnover in the prefrontal cortex and decreases midbrain tryptophan hydroxylase-2 expression in adult female rats: dissociation from estrogenic effects. J Psychiatr Res 43, 656-663.

83 Moriguchi T, Greiner RS, Salem N Jr (2000) Behavioral deficits associated with dietary induction of decreased brain docosahexaenoic acid concentration. J Neurochem 75, 2563-2573.

84 Wainwright PE (2002) Dietary essential fatty acids and brain function: a developmental perspective on mechanisms. Proc Nutr Soc 61, 61-69.

85 Greiner RS, Moriguchi T, Slotnick BM, Hutton A, Salem N (2001) Olfactory discrimination deficits in n3 fatty acid-deficient rats. Physiol Behav 72, 379-385.

86 Catalan J, Moriguchi T, Slotnick B, Murthy M, Greiner RS, Salem NJ (2002) Cognitive deficits in docosahexaenoic acid-deficient rats. Behav Neurosci 116,1022-1031.

87 Salem N Jr, Litman B, Kim HY, Gawrisch K (2001) Mechanisms of action of docosahexaenoic acid in the nervous system. Lipids 36, 945-959.

88 Jeffrey BG, Mitchell DC, Gibson RA, Neuringer M (2002) n-3 fatty acid deficiency alters recovery of the rod photoresponse in rhesus monkeys. Invest Ophthalmol Vis Sci 43, 2806-2814.

89 Jeffrey BG, Mitchell DC, Hibbeln JR, Gibson RA, Chedester AL, Salem N (2002) Visual acuity and retinal function in infant monkeys fed long-chain PUFA. Lipids 37, 839–848.

90 Barinaga M (1998) New lead to brain neuron regeneration. Science 282, 1018-1019.

91 Doidge N (2007) The Brain That Changes Itself. New York, Viking Press.

92 Ahmad A, Murthy M, Greiner RS, Moriguchi T, Salem N Jr (2002) A decrease in cell size accompanies a loss of docosahexaenoate in the rat hippocampus. Nutr Neurosci 5, 103-113

93 Takeuchi T, Fukumoto Y, Harada E (2002) Influence of a dietary n−3 fatty acid deficiency on the cerebral catecholamine contents, EEG and learning ability in rat. Behav Brain Res 131, 193-204.

94 Ximenes da Silva A, Lavialle F, Gendrot G, Guesnet P, Alessandri JM, Lavialle M (2002) Glucose transport and utilization are altered in the brain of rats deficient in n- 3 polyunsaturated fatty acids. J Neurochem 81, 1328-1337.

Chapter 1. Introduction

95 Nanjo A, Kanazawa A, Sato K, Banno F, Fujimoto K (1999) Depletion of dietary n- 3 fatty acid affects the level of cyclic AMP in rat hippocampus. J Nutr Sci Vitaminol 45, 633-641.

96 Gazzah N, Gharib A, Croset M, Bobillier P, Lagarde M, Sarda N (1995) Decrease of brain phospholipid synthesis in free-moving n-3 fatty acid deficient rats. J Neurochem 64, 908-918.

97 Murthy M, Hamilton J, Greiner RS, Moriguchi T, Salem N, Kim HY (2002) Differential effects of n-3 fatty acid deficiency on phospholipid molecular species composition in the rat hippocampus. J Lipid Res 43, 611-617.

98 Cowing BE, Saker KE (2001) Polyunsaturated fatty acids and epidermal growth factor receptor/mitogen-activated protein kinase signaling in mammary cancer. J Nutr 131, 1125-1128.

99 Calder PC, Yaqoob P, Thies F, Wallace FA, Miles EA (2002) Fatty acids and lymphocyte functions. Br J Nutr 87, S31-S48.

100 Calder PC, Grimble RF (2002) Polyunsaturated fatty acids, inflammation and immunity. Eur J Clin Nutr 56(Suppl. 3), S14–S19.

101 Serhan CN, Savill J (2005) Resolution of inflammation: the beginning programs the end. Nat Immunol 6, 1191-1197.

102 Bazan NG (2007) Homeostatic regulation of photoreceptor cell integrity: significance of the potent mediator neuroprotectin D1 biosynthesized from docosahexaenoic acid: the Proctor Lecture. Invest. Ophthalmol. Vis Sci 48, 4866- 4881.

103 Komprda T (2012) Eicosapentaenoic and docosahexaenoic acids as inflammation- modulating and lipid homeostasis influencing nutraceuticals: A review. J Functional Foods 4, 25-38.

104 Rakhshandehroo M, Knoch B, Müller M, Kersten S (2010)Peroxisome proliferator- activated receptor alpha target genes. PPAR Research. doi:10.1155/2010/612089.

105 Shimano H (2001) Sterol regulatory element-binding proteins (SREBPs): transcriptional regulators of lipid synthetic genes (Review). Prog Lipid Res 40, 439- 452.

106 Mietus-Snyder M, Krauss RM (2008) Lipid metabolism in children and adolescents: impact on vascular biology. J Clin Lipidol 2, 127-137.

107 Niinikoski H, Lagström H, Jokinen E, Siltala M, Rönnemaa T, Viikari J, Raitakari OT, Marniemi J, Näntö-Salonen K, Simell O (2007) Impact of repeated dietary counselling between infancy and 14 years of age on dietary intakes and serum lipids and lipoproteins: the STRIP study. Circulation 116, 1032-1040.

Chapter 1. Introduction

108 Russo GL (2009) Dietary n-6 and n-3 polyunsaturated fatty acids: from biochemistry to clinical implication for cardiovascular prevention. Biochem Pharm 77, 937-946.

109 Riediger ND, Othman RA, Suh M, Moghadasian MH (2009) A Systemic Review of the Roles of n-3 Fatty Acids in Health and Disease. J Am Diet Assoc 109, 668-679.

110 Moyad MA (2005) An introduction to dietary/supplemental omega-3 fatty acids for general health and prevention: Part I. Urologic Oncology 23, 28-35.

111 Moyad MA (2005) An introduction to dietary/supplemental omega-3 fatty acids for general health and prevention: Part II. Urologic Oncologic 23, 36-48.

112 Dangardt F, Osika W, Chen Y, Nilsson U, Gan LM, Gronowitz E et al (2010) Omega-3 fatty acid supplementation improves vascular function and reduces inflammation in obese adolescents. Atherosclerosis 212, 580-585.

113 Hawthorne KM, Abrams SA, Heird WC (2009) Docosahexaenoic acid (DHA) supplementation of orange juice increases plasma phospholipid DHA content of children. J Am Diet Assoc 109, 708-712.

114 López-Huertas E (2010) Health effects of oleic acid and long chain omega-3 fatty acids (EPA and DHA) enriched milks. A review of intervention studies, Pharmacological Res 61, 200-207.

115 Ridker PM (2003) Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation 107, 363-369.

116 Klein-Platat C, Drai J, Oujaa M, Schlienger J, Simon C (2005) Plasma fatty acid composition is associated with the metabolic syndrome and low-grade inflammation in overweight adolescents. Am J Clin Nutr 82, 1178-1184.

117 Harris WS, Von Schacky C (2004) The omega-3 index: a new risk factor death from coronary heart disease? Prev Med 39, 212-220.

118 Burrows T, Collins CE, Garg ML (2011) Omega-3 index, obesity and insulin resistance in children. Int J Pediatr Obes 6, e532-e539.

119 Harris WS (2008) The omega-3 index as a risk factor for coronary heart disease. Am J Clin Nutr 87(suppl), 1997S-2002S.

120 von Schacky C (2011) The Omega-3 Index as a risk factor for cardiovascular diseases. Prostaglandins Other Lipid Mediat 96, 94-98.

121 O’Sullivan TA, Ambrosini GL, Mori TA, Beilin LJ, Oddy WH (2011) Omega-3 index correlates with healthier food consumption in adolescents and with reduced cardiovascular disease risk factors in adolescent boys. Lipids 46, 59-67.

122 Saito E, Okada T, Abe Y, Kuromori Y, Miyashita M, Iwata F, Hara M, Ayusawa M, Mugishima H, Kitamura Y (2011) Docosahexaenoic acid content in plasma

Chapter 1. Introduction

phospholipids and desaturase indices in obese children. J Atheroscler Thromb 18, 345-350.

123 Romeo J, Wärnberg J, García-Mármol E, Rodrı´guez-Rodrı´guez M, Diaz LE, Gomez-Martı´nez S et al (2011) Daily consumption of milk enriched with fish oil, oleic acid, minerals and vitamins reduces cell adhesion molecules in healthy children. Nutr Metab Cardiovasc Dis 21, 113-120.

124 Ross R (1993) The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 362, 801-809.

125 Roldán V, Marín F, Lip GYH, Blann AD (2003) Soluble E-selectin in cardiovascular disease and its risk factors: a review of the literature. Thromb Haemost 90, 1007-1020.

126 Engler MM, Engler MB, Malloy M, Chiu E, Besio D, Paul S, Stuehlinger M, Morrow J, Ridker P, Rifai N, Mietus-Snyder M (2004) Docosahexaenoic acid restores endothelial function in children with hyperlipidemia: results from the EARLY Study. Int J Clin Pharmacol Ther 42, 672-679.

127 Mardia L, Araceli M-C, Oscar V-C, Enrique R-M, José F (2011) Supplementation of n3 Long-chain Polyunsaturated Fatty Acid Synergistically Decreases Insulin Resistance with Weight Loss of Obese Prepubertal and Pubertal Children. Archives of Medical Research 42, 502-508.

128 Lemaitre RN, King IB, Mozaffarian D, Kuller LH, Tracy RP, Siscovick DS (2003) N-3 polyunsaturated fatty acids, fatal ischemic heart disease, and nonfatal myocardial infarction in older adults: the Cardiovascular Health Study. Am J Clin Nutr 77, 319-325.

129 He K, Song Y, Daviglus ML, Liu K, Horn LV, Dyer AR, Greenland P ( 2004) Accumulated evidence on fish consumption and coronary heart disease mortality: A meta-analysis of cohort studies. Circulation 109, 2705-2711.

130 Hu F B, Bronner L, Willett WC, Stampfer MJ, Rexrode KM, Albert CM, Hunter D, Manson JE (2002) Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. J Am Med Assn 287, 1815-1821.

131 Iso H, Kobayashi M, Ishihara J, Sasaki S, Okada K, Kita Y, Kukubo Y, Tsugane S (2006) Intake of fish and n−3 fatty acids and risk of coronary heart disease among Japanese: the Japan Public Health Center-Based (JPHC) Study Cohort I. Circulation 113, 195-202.

132 Yokoyama M, Origasa H, Matsuzaki M, Matsuzawa Y, Saito Y, Ishikawa et al. (2007) Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet 369, 1090-1098.

Chapter 1. Introduction

133 Goede J, Geleijnse JM, Boer JMA, Kromhout D, Verschuren WMM (2010) Marine (n-3) fatty acids, fish consumption, and the 10-year risk of fatal and nonfatal coronary heart disease in a large population of Dutch adults with low fish intake. J Nutr 140, 1023-1028.

134 Roche HM (1999) Unsaturated fatty acids. Proceeding Nutr Soc 58, 397-401.

135 Uauy R, Dangour AD (2009) Fat and fatty acid requirements and recommendations for infants of 0-2 years and children of 2-18 years. Ann Nutr Metab 55, 76-96.

136 Meyer BJ, Kolanu N (2011) Australian children are not consuming enough long- chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 27, 1136- 1140.

137 Ervin B, Wright JD, Wang CY, Kennedy-Stephenson J, Division of Health and Nutrition Examination Surveys (2004) Dietary Intake of fats and fatty acids for the United States population: 1999-2000. Advanced data from vital health statistic, CDC Number 348, November 8.

138 Barbarich BN, Willows ND, Wang L, Clandinin MT (2006) Polyunsaturated fatty acids and anthropometric indices of children in rural China. Eur J Clin Nutr 60, 1100-1107.

139 Bermudez OI, Toher C, Montenegro-Bethancourt G, Vossenaar M, Mathias P, Doak C et al (2010) Dietary intakes and food sources of fat and fatty acids in Guatemalan school children: A cross-sectional study. J Nutr 9, 20.

140 Murakami K, Miyake Y, Sasaki S, Tanaka K, Arakawa M (2010) Fish and n-3 Polyunsaturated Fatty Acid Intake and Depressive Symptoms: Ryukyus Child Health Study. Pediatrics 126, e623-e630.

141 Sichert-Hellert W, Wicher M, Kersting M (2009) Age and time trends in fish consumption pattern of children and adolescents, and consequences for the intake of long-chain n-3 polyunsaturated fatty acids. Eur J Clin Nutr 63, 1071-1075.

142 Sioen I, Huybrechts I, Verbeke W, Van Camp J, De Henauw S (2007) n-6 and n-3 PUFA intakes of pre-school children in Flanders, Belgium. Br J Nutr 98, 819-825.

143 Australian Government, Department of Health and Ageing, National Health and Medical Research Council (2005) Food for health, dietary guidelines for Australian, a guide to healthy eating. Available at: http://www.nhmrc.gov.au /_files_nhmrc/publications/attachments/n31.pdf?q=publications/synopses/_files/n31 .pdf, accessed on February, 12, 2012.

144 National Health and Medical Research Council (2013) Australian Dietary Guidelines. Canberra: National Health and Medical Research Council. Available at: http://www.nhmrc.gov.au/guidelines/publications/n55, accessed on March, 25, 2013.

Chapter 1. Introduction

145 Dietitians Association of Australia (2011) A modeling system to inform the revision of the Australian guide to healthy eating. Commonwealth of Australia. Available at: http://www.nhmrc.gov.au/_files_nhmrc/file/publications/n55c_australian_dietary_g uidelines_food_modelling.pdf, accessed on March, 25, 2013.

146 Colquhoun D, Ferreira-Jardim A, Udell T, Eden B, the Nutrition and Metabolism Committee of the Heart Foundation. (2008) Review of evidence fish, fish oils, n-3 pollyunsaturated fatty acids and cardiovascular health. Available at: www.heartfoundation.org.au, accessed on December 2011.

147 National Health Foundation of Australia (2008) Position statement, fish, fish oils, n- 3 polyunsaturated fatty acids and cardiovascular health. Available at: http://www.heartfoundation.org.au/sitecollectiondocuments/fish-position- statement.pdf, accessed on January, 15, 2011.

148 National Health Medical Research Council (2006) Nutrient reference value for Australia and New Zealand including recommended dietary intakes. Australian Government Department of Health and Ageing. Commonwealth of Australia. Available at: http://www.nhmrc.gov.au/_files_nhmrc/publications/attachments /n35. pdf, accessed on June, 6, 2010

149 Leek S, Muddock S, Foxall G (2000) Situational determinants of fish consumption. Br Food J 102, 18-39.

150 Trondsen T, Scholderer J, Lund E, Eggen AE (2003) Perceived barriers to consumption of fish among Norwegian women. Appetite 41, 301-314.

151 Verbeke W, Vackier I (2005) Individual determinants of fish consumption: application of the theory of planned behaviour. Appetite 44, 67-82.

152 Mozaffarian D, Rimm EB (2006) Fish intake, contaminants, and human health evaluating the risks and the benefits. J Am Med Assoc 296, 1885-1899.

153 Food and Agriculture Organization/World Health Organization (2011) Report of the joint FAO/WHO expert consultation on the risk and benefits of fish consumption, 25-29 January 2010, Rome, Italy. FAO Fisheries and Aquaculture Report, No 978.

154 English R, Cashel KM, Lewis JL, Bennet S, Berzins J, Penm R (1988) National Dietary Survey of Schoolchildren aged 10–15 Years. Report no. 1. Foods Consumed. Canberra: Australian Government Publishing Service.

155 McLennan W, Podger A (1999) National Nutrition Survey – Foods Eaten, Australia. Canberra: Australian Bureau of Statistics, Commonwealth Department of Health and Aged Care.

156 Fisheries Research and Development Corporation (1998) A study of the retail sale and in home consumption of seafood in Sydney. Canberra, Commonwealth Dept of Agriculture, Fishery and Forestry.

Chapter 1. Introduction

157 Fisheries Research and Development Corporation (1999) A study of seafood consumption in Perth. Canberra, Commonwealth Dept of Agriculture, Fishery and Forestry.

158 Fisheries Research and Development Corporation (2011) Community perceptions of the sustainability of the fishing industry in Australia.

159 Howe P, Buckley J, Meyer B (2007) Red meat: a source of long chain omega-3. Nutr Diet 64, S135-139.

160 Clayton EH, Hanstock TL, Watson JF (2009) Estimated intakes of meat and fish by children and adolescents in Australia and comparison with recommendations. Br J Nutr 101, 1731-1735.

161 O’Sullivan BL, Williams BG, Meyer BJ (2006) Biomarker Validation of a Long- Chain Omega-3 Polyunsaturated Fatty Acid Food Frequency Questionnaire. Lipids 41, 845-850.

162 Sullivan BL, Brown J, Williams PG, Meyer BJ (2008) Dietary validation of a new food frequency questionnaire that estimates long-chain omega-3 polyunsaturated fatty acids. Br J Nutr 99, 660-666.

163 Swierk M, Williams PA, Wilcox J, Russell KG, Meyer BJ (2011) Validation of an Australian electronic food frequency questionnaire to measure polyunsaturated fatty acid intake. Nutrition 27, 641-646.

164 Zak L (2008) Breaking bread habits of Australia’s manufacturers. Food magazine. Available at: http://www.foodmag.com.au/news/breaking-bread-habits-of-australia- s-manufacturers, accessed on March, 1, 2013.

165 Patch CS, Tapsell LC, Mori TA, Meyer BJ, Murphy KJ, Mansour J et al. (2005) The use of novel foods enriched with long chain n-3 fatty acids to increase dietary intake: a comparison of methodologies assessing nutrient intake. J Am Diet Assoc 105, 1918-1926.

166 Garg ML, Wood LG, Singh H, Moughan PJ (2006) Means of delivering recommended levels of long chain n-3 polyunsaturated fatty acids in human diets. J Food Sci 71, R66-R71.

167 Prell H, Berg C, Jonsson L (2002) Why don’t adolescents eat fish? Factors influencing fish consumption in school. Scandinavian J Articles 46, 184-191.

168 Pliner P, Salvy SJ. Food neophobia in humans (2006) In R Shepherd , M Raats (Eds). The psychology of food choice. Oxfordshire, CABI, 75-92.

169 Knaapila A, Silventoinen K, Broms U, Rose RJ, Perola M, Kaprio J, Tuorila HM (2011) Food neophobia in young adults: genetic architecture and relation to personality, pleasantness and use frequency of foods, and body mass index – A twin study. Behav Genet 41, 512-521.

Chapter 1. Introduction

170 Li D, Sinclair A, Mann N, Turner A, Kelly F, Abedin L, Wilson A (1999) Effect of dietary α-linolenic acid on thrombotic risk factors in vegetarian men. Am J Clin Nutr 69, 872-882.

171 SACN/COT Scientific Advisory Committee on Nutrition (SACN) and Committee on Toxicity (COT) authors (2004) Advice on Fish Consumption: Benefits and Risks. TSO, Norwich, UK.

172 Murphy KJ, Meyer BJ, Mori TA, Burke V, Mansour J, Patch CS et al. (2007) Impact of food enriched with n-3 long-chain polyunsaturated fatty acids on erythrocyte n-3 levels and cardiovascular risk factors. Br J Nutr 97, 749-757.

173 Mu H (2008) Bioavailability of omega-3 long-chain polyunsaturated fatty acids from foods. Agro Food Industry hi-tech 19, 24-26.

174 Kolanowski W (2005) Bioavailability of omega-3 PUFA from foods enriched with fish oil - A mini review. Pol J Food Nutr Sci 14, 335-340.

175 Newton I, Snyder D (1993) Nutritional aspects of long-chain omega-3 fatty acids and their use in bread enrichment. Cereal Foods World 3, 126-131. Available at: http://www.ceresconsulting.com/images/BREAD_Fortification.pdf, accessed on June, 13, 2012.

176 Wallace JMW, McCabe AJ, Robson PJ, Keogh MK, Murray CA, Kelly PM, Marquez-Ruiz G, McGlynn H, Gilmore WS, Strain JJ (2000) Bioavailability of n-3 polyunsaturated fatty acids (PUFA) in food enriched with microencapsulated fish oil. Ann Nutr Metab 44, 157-162.

177 Liu M, Wallin R, Saldeen T (2001) Effect of bread containing stable fish oil on plasma phospholipid fatty acids, triglycerides, HDL-cholesterol and malondialdehyde in subjects with hyperlipidemia. Nutr Res 21, 1403-1410.

178 Church DF, Pryor WA (1985) Free radical chemistry of cigarette smoke and its toxicological implications. Environ Health Perspect 64, 111-126.

179 Nielsen F, Mikkelsen BB, Nielsen JB, Andersen HR, Grandjean P (1997) Plasma malondialdehyde as biomarker for oxidative stress: reference interval and effects of life-style factors. Clinical Chemistry 43, 1209-1214.

180 Yep YL, Li D, Mann NJ, Bode O, Sinclair AJ (2002) Bread enriched with microencapsulated tuna oil increases plasma docosahexaenaoic acid and total omega-3 fatty acids in humans. Asia Pac J Clin Nutr 11, 285-291.

181 Grune T, Kramer K, Hoppe PP, Siems W (2001) Enrichment of eggs with n-3 polyunsaturated fatty acids: effects of vitamin E supplementation. Lipids 36, 833- 838.

Chapter 1. Introduction

182 Lopez-Ferrer S, Baucells MD, Barroeta AC, Grashorn MA (2001) n-3 enrichment of chicken meat. 1. Use of very long-chain fatty acids in chicken diets and their influence on meat quality: fish oil. Poult Sci 80, 741-752.

183 Ayerza R, Coates W (2000) Dietary Levels of Chia: Influence on Yolk Cholesterol, Lipid Content and Fatty Acid Composition for Two Strains of Hens. Poultry Science 79, 724-739.

184 Jiang Z, Sim JS (1993) Consumption of n-3 polyunsaturated fatty acid-enriched eggs and changes in plasma lipids of human subjects. Nutrition 9, 513-518.

185 Ferrier LK, Caston LJ, Leeson S, Squires J, Weaver BJ, Holub BJ (1995) α- Linolenic acid-and docosahexaenoic acid-enriched eggs from hens fed flaxseed: influence on blood lipids and platelet phospholipid fatty acids in humans. Am J Clin Nutr 62, 81-86.

186 Delaroudis S, Slavakis A, Kyroudi A, Tzoiti, Sarris, Balaris et al. (1999) Omega-3 PUFA modified eggs: a study on their effect on the lipid profile of human volunteers. 71st EAS’ Congress and Satellite Symposia, 130-131.

187 Lewis NM, K Schalch, Scheideler (2000) Serum lipid response to n-3 fatty acid enriched eggs in persons with hypercholesterolemia. J Am Diet Assoc 100, 365-367.

188 Gillingham LG, Caston L, Leeson S, Hourtovenko K, Holub BJ (2005) The effects of consuming docosahexaenoic acid (DHA)-enriched eggs on serum lipids and fatty acid composition in statin-treated hypercholesterolemic male patients. Food Res International 38, 1117-1123.

189 Öhman M, Åkerfeldt T, Nilsson I, Rosen C, Hansson LO, Carlsson M, Larsson A (2008) Biochemical effects of consumption of eggs containing omega-3 polyunsaturated fatty acids. Upsala J Med Sci 113, 315-323.

190 Walldius G, Jungner I (2004) Apolipoprotein B and apolipoprotein A-I: risk indicators of coronary heart disease and targets for lipid-modifying therapy. J Intern Med 255, 188-205.

191 Savas Erdeve S, Simsek E, Dallar Y, Biyikli Z (2010) Utility of ApoB/ApoA1 ratio for the prediction of cardiovascular risk in children with metabolic syndrome. Indian J Pediatr 77, 1261-1265.

192 Nolan JM, Stack J, O’Connell E, Beatty S (2007) The relation between macular pigment optical density and its constituent carotenoids in diet and serum. Invest Ophthalmol Vis Sci 48, 571-582.

193 Surai P, MacPherson A, Speake BK, Sparks NH (2000) Designer egg evaluation in a controlled trial. Eur J Clin Nutr 54, 298-305.

Chapter 1. Introduction

194 Burns-Whitmore BL, Haddad EH, Sabate J, Jaceldo-Siegl K, Tanzman J, Rajaram S (2010) Effect of n-3 fatty acid enriched eggs and organic eggs on serum lutein in free-living lacto-ovo vegetarians. Eur J Clin Nutr 64, 1332-1337.

195 Johnson EJ, Chung H-Y, Caldarella SM, Snodderly DM (2008) The influence of supplemental lutein and docosahexaenoic acid on serum, lipoproteins, and macular pigmentation. Am J Clin Nutr 87, 1521-1529.

196 Visioli F, Rise P, Plasmati E, Pazzucconi F, Sirtori CR, Galli C (2000) Very low intakes of n-3 fatty acids incorporated into bovine milk reduce plasma triacylglycerol and increase HDL-cholesterol concentrations in healthy subjects. Pharm Res 41, 571-576.

197 Higgins S, Carroll YL, O’Brien NM, Morrissey A (1999) Use of microencapsulated fish oil as a mean of increasing n-3 polyunsaturated fatty acid intake. J Hum Nutr Diet 12, 265-271.

198 Baró L, Fonollá J, Peńa JL, Martínez-Férez A, Lucena A, Jiménez J, Boza JJ, López-Huertas E (2003) n-3 fatty acids plus oleic acid and vitamin supplemented milk consumption reduces total and LDL cholesterol, hemocysteine and level of endothelial adhesion molecules in healthy humans. Clin Nutr 22, 175-182.

199 Ohta T, Saku K, Takata K, Adachi N (1999) Soluble vascular cell-adhesion molecule-1 and soluble intercellular adhesion molecule-1 correlate with lipid and apolipoprotein risk factors for coronary artery disease in children. Eur J Pediatr 158, 592-598.

200 Blankenberg S, Rupprecht HJ, Bickel C, Jiang XC, Poirier O, Lackner KJ, Meyer J, Cambien F, Tiret L (2003) Common genetic variation of the cholesteryl ester transfer protein gene strongly predicts future cardiovascular death in patients with coronary artery disease. J Am Coll Cardiol 41, 1983-1989.

201 Thienprasert A, Samuhaseneetoo S, Popplestone K, West AL, Miles EA, Calder PC (2009) Fish Oil N-3 Polyunsaturated Fatty Acids Selectively Affect Plasma Cytokines and Decrease Illness in Thai Schoolchildren: A Randomized, Double- Blind, Placebo-Controlled Intervention Trial. J Pediatrics 154, 391-395.

202 Jacobsen C, Let MB, Andersen G, Meyer AS (2006) Oxidative stability of fish oil enriched yoghurts. In Seafood research from fish to dish: Ouality, safety and processing of wild and farmed fish, 1st ed. Luten JB, Jaconsen C, Bakaert K, Sæbø A, Oehlenschläger J, Eds. Wageningen Academic Publishers: Wageningen, The Nederlands.

203 Let MB, Jacobsen C, Meyer (2007) Lipid oxidation in milk, yoghurt, and salad dressing enriched with neat fish oil or pre-emulsified fish oil. J Agric Food Chem 55, 7802-7809.

Chapter 1. Introduction

204 Farvin KH, Baron CP, Nielsen NS, Jacobsen C (2010) Antioxidant activity of yoghurt peptides: Part 1-in vitro assays and evaluation in ω-3 enriched milk. Food Chemistry 123, 1081-1089.

205 Schram LB, Nielsen CJ, Porsgaard T, Nielsen NS, Holm R, Mu H (2007) Food matrices affect the bioavailability of (n-3) polyunsaturated fatty acids in a single meal study in humans. Food Res Inter 40, 1062-1068.

206 Garg ML, Blake RJ, Clayton E, Munro IA, Macdonald-Wicks L, Singh H, Moughan PJ (2007) Consumption of an n-3 polyunsaturated fatty acid-enriched dip modulates plasma lipid profile in subjects with diabetes type II. Eur J Clin Nutr 61, 1312-1317.

207 Arterburn LM, Oken HA, Bailey Hall E, Hamersley J, Kuratko CN, Hoffman JP (2008) Algal-oil capsules and cooked salmon: nutritionally equivalent sources of docosahexaenoic acid. J Am Diet Assoc 108, 1204-1209.

208 Brug J, Oenema A, Ferreira I (2005) Theory, evidence and Intervention Mapping to improve behavior nutrition and physical activity interventions. Int J Behav Nutr Phys Act 2, 2, doi: 10.1186/1479-5868-2-2.

209 McKenzie JF, Smeltzer JL (2001) Planning, implementing, and evaluating health promotion programs: A primer. Third edition edition. Boston, Allyn and Bacon; 2001.

210 Bartholomew LK, Parcel GS, Kok G (1998) Intervention mapping: A process for developing theory and evidence-based health education programs. Health Educ Behav 25, 545-563.

211 Department of Health. Review of children's healthy eating interventions, 2011. Available at: http://docs.health.vic.gov.au/docs/doc/Full-Review-of-Childrens- Healthy-Eating-Interventions, accessed on January, 25, 2013.

212 Bartholomew LK, Parcel GS, Kok G, Gottlieb NH (2006) Planning health promotion programs: an Intervention Mapping approach (2nd ed.). San Francisco, CA, Jossey-Bass.

213 Bartholomew LK, Parcel GS, Kok G, Gottlieb NH, Fernández, ME (2011) Planning health promotion programs; an Intervention Mapping approach, 3rd Ed. San Francisco, CA, Jossey-Bass.

214 Baranowski T, Weber Cullen K, Baranowski J (1999) Psychosocial correlates of dietary intake: Advancing dietary intervention. Annu Rev Nutr 19, 17-40.

215 Kok G, Lo SH, Peters G-JY, Ruiter RAC (2011) Changing energy-related behavior: An Intervention Mapping approach. Energy Policy 39, 5280-5286.

216 Cullen KW, Bartholomew LK, Parcel GS, Kok G (1998) Intervention Mapping: Use of theory and data in the development of a nutrition program to increase fruit intake and vegetable intake in girls, ages 9-12. J Nutr Educ 30, 188-195.

Chapter 1. Introduction

217 Perez-Rodrigo C, Wind M, Hildonen C, Bjelland M, Aranceta J, Klepp KI, et al. (2005) The pro children intervention: applying the intervention mapping protocol to develop a school-based fruit and vegetable promotion programme. Ann Nutr Metab 49, 267-277.

218 McManus A, White J, Hunt W, Storey J, McManus J, Cuesta-Briand B, Golightly A (2011) Community intervention to increase seafood consumption (CIISC). Centre of Excellence for Science Seafood & Health (CESSH), Curtin Health Innovation.

219 Corbie-Smith G, Akers A, Blumenthal C, Council B, Wynn M, Muhammad M, Stith D (2010) Intervention Mapping as a Participatory Approach to Developing an HIV prevention Intervention in Rural African American Communities. AIDS Educ Prev 22, 184-202.

220 Birch LL (1979) Preschool children’s food preferences and consumption patterns. J Nutr Educ 11, 189-192.

221 Skinner JD, Carruth BR, Bounds W, Ziegler PJ (2002) Children’s food preferences: A longitudinal analysis. J Am Diet Assoc 102, 1638-1647.

222 Kral TVE, Rauh EM (2010) Eating behaviour of children in the context of their family environment. Phy and Behav 100, 567-573.

223 Savage JS, Fisher JO, Birch LL (2007) Parental influence on eating behaviour: conception to adolescence. J Law Med Ethics 35, 22-34.

224 Imm P, Knobeloch L, Anderson HA (2006) Maternal recall of children’s consumption of commercial and sport-caught fish: Findings from a multi-state study. Environ Res 103, 198-204.

225 Birch LL (1999) Development of food preferences. Annu Rev Nutr 9, 41-62.

226 Gilman M, Rifas-Shiman S, Frazier L et al. (2000) Family dinner and diet quality among older children and adolescents. Archives Family Med 9, 235-240.

227 Jebb S, Steer T, Holmes C (2007) The ‘healthy living’ social marketing initiative: A review of the evidence. Available at: http://www.mrc-hnr.cam.ac.uk/wordpress/wp- content/uploads/DH_Healthy-Living-Social-Marketing_0730521.pdf, accessed on April, 20, 2013.

228 Jenkin S, Horner SD (2005) Barriers that influence eating behaviours in adolescents. J Paediatr Nurs 20, 258-267.

229 Allen L, Benoist B, Dary O, Hurrell R (2006) Guidelines on food fortification with micronutrients. World Health Organisation, Food and Agricultural Organization of the United Nations. Available at: http://www.who.int/nutrition/publications/guide _food_fortification_micronutrients.pdf, accessed on April, 1, 2012.

Chapter 1. Introduction

230 King JC (2007) An evidence-based approach for establishing dietary guidelines. J Nutr 137, 480-483.

231 Hornick BA, Krester AJ, Nicklas TA (2008) Menu Modeling with Mypyramid Food Patterns: Incremental Dietary Changes Lead to Dramatic Improvements in Diet Quality of Menus. J Am Diet Assoc 108, 2077-2083.

232 Britten P, Lyon J, Weaver CM, Kris-Etherton PM, Nicklas TA, Weber JA, Davis CA (2006) MyPyramid Food Intake Pattern Modeling for the Dietary Guidelines Advisory Committee. J Nutr Edu Behav 38, S143-S152.

233 Newby PK, Muller D, Hallfrisch J, Andres R, Tucker KL (2004) Food patterns measured by factor analysis and antropometric changes in adults. Am J Clin Nutr 80, 504-513.

234 Hu FB (2002) Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 13, 3-9.

235 Fung TT, Schulze M, Manson JE, Willett WC, Hu FB (2004) Dietary patterns, meat intake, and the risk of type 2 diabetes in women. Arch Intern Med 164, 2235-2240.

236 Reedy J, Wirfält E, Flood A, Mitrou PN, Krebs-Smith SM, Kipnis V et al. (2010) Comparing 3 dietary pattern methods-cluster analysis, factor analysis, and index analysis-with colorectal cancer tisk. The NIH-AARP diet and health study. Am J Epidemol 171, 479-487.

237 Venkaiah K, Brahmam GNV, Vijayaraghavan K (2011) Application of factor analysis to identify dietary patterns and use of factor scores to study their relationship with nutritional status of adult rural populations. J Health Popul Nutr 4, 327-338.

238 Moeller SM, Reedy J, Millen AE, Dixon LB, Newby PK, Tucker KL et al . (2007) Dietary patterns: challenges and opportunities in dietary patterns research an experimental biology workshop, April 1, 2006. J Am Diet Assoc 107, 1233-1239.

239 Stricker MD, Onland-Moret NC, Boer JMA, van der Schouw YT, Verschuren WMM, May AM, Peeters PHM, Beulens JWJ (2013) Dietary patterns derived from principal component-and k-means cluster analysis: Long-term association with coronary heart disease and stroke. Nutr Metab Cardio Diseases 23, 250-256.

240 Lioret S, McNaughton SA, Crawford D et al. (2011) Parents’ dietary patterns are significantly correlated: findings from the Melbourne Infant Feeding Activity and Nutrition Trial (InFANT) Program. Br J Nutr 108, 518–526.

241 Subar AF, Crafts J, Zimmerman TP, Wilson M, Mittl B, Islam NG et al. (2010) Assessment of the accuracy of portion size reports using computer-based food photographs aids in the development of an automated self-administered 24-hour recall. J Am Diet Assoc 110, 55-64.

Chapter 1. Introduction

242 Margetts, Nelson (1997) Design and concepts in nutritional epidemiology. Oxford University Press.

243 Haraldsottir J, Holm L, Astrup AV, Halkjoer J, Stender S (2001) Monitoring of dietary changes by telephone interviews: results from Denmark. Public Health Nutrition 4,1287-1295.

244 Willett WC. Nutritional Epidemiology, Second Edition. New York: Oxford University Press, 1998

245 Matos SM, Prado MS, Santos CA, D'Innocenzo S, Assis AM, Dourado LS, Oliveira NS, Rodrigues LC, Barreto ML (2012) Validation of a food frequency questionnaire for children and adolescents aged 4 to 11 years living in Salvador, Bahia. Nutr Hosp 4, 1114-1119.

246 Bel-Serrat S, Mouratidou T, Pala V, Huybrechts I, Börnhorst C, Fernández-Alvira JM, Hadjigeorgiou C, Eiben G, Hebestreit A, Lissner L, Molnár D, Siani A, Veidebaum T, Krogh V, Moreno LA (2013) Relative validity of the Children's Eating Habits Questionnaire-food frequency section among young European children: the IDEFICS Study. Public Health Nutr 4, 1-11.

247 Jarman M, Fisk CM, Ntani G, Crozier SR, Godfrey KM, Inskip HM, Cooper C, Robinson SM, the Southampton Women's Survey Study Group (2013) Assessing diets of 3-year-old children: evaluation of an FFQ. Public Health Nutr 1, 1-9.

248 Kwok FY, Ho YY, Chow CM, So CY, Leung TF (2013) Assessment of nutrient intakes of picky-eating Chinese preschoolers using a modified food frequency questionnaire. World J Pediatr 9, 58-63.

249 Burrows TL, Truby H, Morgan PJ, Callister R, Davies PS, Collins CE (2012) A comparison and validation of child versus parent reporting of children's energy intake using food frequency questionnaires versus food records: Who's an accurate reporter? Clin Nutr, S0261-5614.

250 Sharma S, Cade J, Jackson M, Mbanya JC, Chungong S, Forrester et al. (1996) Development of food frequency questionnaires in three population samples of African origin from Cameroon, Jamaica and Caribbean migrants to the UK. Eur J Clin Nutr 50, 479-486.

251 Kaaks R, Ferrari P (2006) Dietary intake assessments in epidemiology: can we know what we are measuring?, Ann Epidemiol16, 377-380.

252 Block G (1982) A review of validations of dietary assessment methods. Am J Epidemiol 115, 492-505.

253 Kaaks R (1997) Biochemical markers as additional measurements in studies of the accuracy of dietary questionnaire measurements: conceptual issues, Am J Clin Nutr 65(suppl.), l232S-1239S.

Chapter 1. Introduction

254 Ocke MC, Kaaks RJ (1997) Biochemical markers as additional measurements in dietary validity studies: application of the method of triads with examples from the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr 65, 1240S-1245S.

255 Mikkelsen T, Osler M, Olsen S (2006) Validity of protein, retinol, folic acid and n-3 fatty acid intakes estimated from the food-frequency questionnaire used in the Danish National Birth Cohort. Public Health Nutr 9, 771-778.

256 Mina K, Fritschi L, Knuiman M (2007) A valid semi quantitative food frequency questionnaire to measure fish consumption. Eur J Clin Nutr 61, 1023-1031.

257 McNaughton S, Hughes M, Marks G (2007) Validation of a FFQ to estimate the intake of PUFA using plasma phospholipid fatty acids and weighed foods records. Br J Nutr 97, 561-568.

258 Orton HD, Szabo NJ, Clare-Salzler M, Norris M (2008) Comparison between omega-3 and omega-6 polyunsaturated fatty acid intakes as assessed by a food frequency questionnaire and erythrocyte membrane fatty acid composition in young children. Eur J Clin Nutr 62, 733-738.

259 Zhang B, Wang P, Chen C, He QQ, Zhuo SY, Chen YM, Su XY (2010) Validation of an FFQ to estimate the intake of fatty acids using erythrocyte membrane fatty acids and multiple 3d dietary records. Public Health Nutr 13, 1546-1552.

260 Nelson M, Atkinson M, Darbyshire S (1994) Food photography I: The perception of food portion size from photographs. Br J Nutr 72, 649-663.

261 Baranowski T, Domel SB (1994) A cognitive model of children’s reporting of food intake. Am J Clin Nutr 59 (suppl), 212S-217S.

262 Foster E, Matthews JN, Nelson M, Harris JM, Mathers JC, Adamson AJ (2006) Accuracy of estimates of food portion size using food photographs - the importance of using age-appropriate tools. Public Health Nutr 9, 509-514.

263 Foster E, Matthews JNS, Lloyd J, Marshall L,Mathers JC, Nelson M, Barton KL, Wrieden WL, Cornelissen P, Harris J, Adamson AJ (2008) Children’s estimates of food portion size: the development and evaluation of three portion size assessment tools for use with children. B J Nutr 99, 175-184.

264 Weber JC, Lamb DR (1970) Statistics and Research in Physical Education. St. Louis, CV Mosby Co, 59-64.

Chapter 2. Omega-3 LCPUFA intake study

Chapter 2

The majority of this section is the substantive content of the work published in Lipids2013, 48: 869-877.

The findings of this study were also presented in the 34th Nutrition Society of Australia, Perth 2010 (oral presentation). Abstract was published in the Proceedings of the Nutrition Society of Australia 34, 6.

Statement

As the primary supervisor, I, Barbara J Meyer, declare that the greater part of the work in this paper is attributed to the candidate, Setyaningrum Rahmawaty.

In the manuscript, Setyaningrum contributed to study design and was primarily responsible for data collection, data analysis, data interpretation and writing up the manuscript.

A/Prof Barbara J Meyer (Main Supervisor)

Chapter 2. Omega-3 LCPUFA intake study

Dietary intake and food sources of EPA, DPA and DHA in Australian children

(Published in Lipids 2013, 48: 869-877)

Setyaningrum Rahmawaty1,2, Karen Charlton2, Philippa Lyons-Wall3 and Barbara J Meyer1,2*

1Metabolic Research Centre, University of Wollongong: 2School of Health Sciences, University of Wollongong, Northfields Ave, Wollongong NSW 2522, Australia: 3 School of Exercise and Health Sciences, Edith Cowan University, 270 Joondalup Drive, Joondalup, WA 6027, Australia

Abstract

Secondary analysis of the 2007 Australian National Children’s Nutrition and Physical Activity survey was undertaken to assess the intake and food sources of EPA, DPA and DHA (excluding supplements) in 4,487 children aged 2-16 years. An average of two 24-hour dietary recalls was analysed for each child and food sources of EPA, DPA and DHA were assessed using the Australian nutrient composition database called AUSNUT 2007. Median (inter quartile range, IQR) for EPA, DPA and DHA intakes (mg/d) for 2-3y, 4-8y, 9-13y, 14-16y were: EPA 5.3 (1.5-14), 6.7 (1.8-18), 8.7 (2.6-23), 9.8 (2.7-28) respectively; DPA 6.2 (2.2-14), 8.2 (3.3-18), 10.8 (4.3-24), 12.2 (5-29) respectively; and DHA 3.9 (0.6-24), 5.1 (0.9-26), 6.8 (1.1-27), 7.8 (1.5-33) respectively. Energy-adjusted intakes of EPA, DPA and DHA in children who ate fish were 7.5, 2 and 16-fold higher, respectively (P<0.001) compared to those who did not eat fish during the two days of the survey. Intake of total long chain n-3 PUFA was compared to the energy adjusted suggested dietary target (SDT) for Australian children and 20% of children who ate fish during the two days of the survey met the SDT. Fish and seafood products were the largest contributors to DHA (76%) and EPA (59%) intake, while meat, poultry and game contributed to 56% DPA. Meat was consumed 8.5 times in a greater amount than fish/seafood. Australian children do not consume the recommended amounts of long chain omega-3 fatty acids, especially DHA, which could be explained by low fish consumptions. Key words: EPA, DPA and DHA intake ∙ fish consumption ∙ children ∙ Australia ∙ fatty acids

Corresponding author: [email protected], Ph: +61 (0)2 4221 3459, Fax: +61 (0)2 4221 5945

Chapter 2. Omega-3 LCPUFA intake study

Abbreviations

ALA Alpha-linolenic acid (18:3n-3) DHA Docosahexaenoic acid (22:6n-3) DPA Docosapentaenoic acid (22:5n-3) EPA Eicosapentaenoic acid (20:5n-3) PUFA Polyunsaturated fatty acid(s)

Introduction

Long chain n-3 PUFA comprises of EPA, DPA and DHA, and these nutrients, especially DHA are essential for supporting the normal brain and cognitive development of children [1, 2] and several lines of evidence demonstrate the potential role of these fatty acids to reduce cardiovascular disease (CVD) risk in children [3-6]. Reductions in intake of the long chain n-3 PUFA have been hypothesized as one of the dietary factors associated with the increase in incidence of chronic diseases in Western countries over the past 50 years [7]. Long chain n-3 PUFA are mainly obtained preformed from dietary sources such as fish, with minimal de novo synthesis from elongation of the shorter chain n-3 fatty acid, ALA [8, 9, 10].

Evidence is increasing regarding the beneficial effects of early intake of long chain n-3 PUFA for the prevention of CVD in later life [11, 12]. A low concentration of long chain n-3 PUFA, together with an increased concentration of C-reactive protein (CRP), a sensitive predictor of CVD risk [13], has been found in overweight teenagers [14]; and a low omega-3 index, the sum of EPA and DHA in erythrocyte membranes expressed as a percentage of total fatty acids [15], has been found in obese school age children in association with insulin resistance [16]. Conversely, an increase in long chain n-3 PUFA intake in obese adolescents has been shown to modulate vascular function and inflammatory markers, including lymphocytes, monocytes and levels of tumor necrosis factor -α, interleukin-1β and interleukin-6 [4]. Long chain n-3 PUFA supplementation in healthy children aged 8-14 years has resulted in a reduction in biomarkers of CVD risk, including E-selectin and intracellular adhesion molecules-1 (ICAM-1) [3]. E-selectin and ICAM-1 are members of cell adhesion molecules that are expressed in the endothelium that are involved in the inflammatory development, such as seen in people with CVD [17, 18]. Additionally, DHA supplementation in children

Chapter 2. Omega-3 LCPUFA intake study

with hyperlipidemia at risk of early heart disease has resulted in improved endothelium- dependent flow-mediated dilation (FMD) of the brachial artery [19]. These data suggest that CVD vascular risk factors in children can be reduced by increased long chain n-3 PUFA intake.

In Australia, the National Heart Foundation (NHF) recommends that for cardiovascular health, children should follow the adult recommendation of 500 mg/d of EPA and DHA or at least two, preferably oily, fish meals per week [20]. Other agencies, for example the National Health and Medical Research Council (NHMRC), have established Nutrient Reference Values for long chain n-3 PUFA for children under 14 years, based on the adequate intake (AI) or observed median intake data reported in the 1995 National Dietary Survey as no recommended dietary intake (RDI) for children has been established by the NHMRC [21]. For children aged 14-16 years, a suggested dietary target (SDT) for the prevention of chronic diseases, has been set at 610 and 430 mg/d for boys and girls, respectively, based on the observed 90th percentile of the population intake [21]. Meyer & Kolanu (2011) have extrapolated SDT for long chain n-3 PUFA for children younger than 14 years from adjusted energy intakes, by sex and age group [22] and due to lower energy intake in younger children, these SDT are lower compared to the NHF recommended intakes of 500 mg/d [20].

It has previously been reported that most Australian children do not meet the recommended two fish meals per week [22]. However, limited data are available on individual intakes of the long chain n-3 PUFA, namely EPA, DPA and DHA, and their respective food sources, in both consumers and non-consumers of fish. Therefore the aim of this study was to analyse the 2007 Australian National Children’s Nutrition and Physical Activity survey (Children’s Survey), to determine the average intake (average of two 24-h dietary recalls) of EPA, DPA and DHA as well as their respective food sources in fish and non-fish consumers.

Material and methods

Data

The source of data for this study was the 2007 Children’s Survey conducted between February and August 2007. The Children’s Survey was conducted according to the

Chapter 2. Omega-3 LCPUFA intake study

guidelines laid down in the Declaration of Helsinki and procedures were approved by the NHMRC registered ethics committees of the Commonwealth Scientific and Industrial Research Organization and the University of South Australia [23]. Permission to access the dataset was obtained from the Australian Social Science Data Archive [24]. The dietary intake of 4,487 children aged 2-16 years, selected using random digit dialing (RDD) from all Australian states and territories in metropolitan, rural and remote areas, was assessed from two 24-hour recalls using a standardized multiple pass 24-hour dietary recall methodology. A computer assisted personal interview (CAPI) technique was used to capture the first 24-hr recall, followed by a computer assisted telephone interview (CATI) conducted 7-21 days after the CAPI to obtain the second 24-hour recall. All days of the week are approximately equally represented in the database of the Children’s survey [25]. Details of the survey methodology have been reported previously [25].

Dietary Intake of EPA, DPA and DHA assessment

Dietary intake was obtained from the average of the two 24-hour dietary recalls analysed for intakes of EPA, DPA and DHA from foods consumed (excluding fish oil supplements but including foods enriched with n-3 LCPUFA) using the Australian nutrient composition database called AUSNUT 2007 which was developed specifically for the Children’s survey. In the AUSNUT 2007 database certain foods such as margarine were erroneously allocated long chain n-3 PUFA values; hence these values were corrected and correct values for EPA, DPA and DHA intakes were re-calculated. The intake of total long chain n-3 PUFA was determined by summing individual intakes of EPA, DPA, and DHA and was compared to the energy adjusted SDT for long chain n-3 PUFA for Australian children [22].

Food items were categorized into the following groups as previously reported [25]: non-alcoholic beverages; cereal and cereal products; cereal-based products and dishes; fats and oils; fish and seafood products and dishes; fruit products and dishes; egg products and dishes; meat and poultry and game products and dishes; milk products and dishes; dairy substitutes; soup; seed and nut products and dishes; savory sauces and condiments; vegetable products and dishes; legume and pulse products and dishes; snack foods; sugar products and dishes; confectionery and cereal/nut/fruit/seed bars; alcoholic beverages; special dietary foods (e.g. formula dietary foods and enteral 62

Chapter 2. Omega-3 LCPUFA intake study

formula); miscellaneous (e.g. yeast, herbs, spices, and food enhancer); and infant formulae and foods [25]. The percentage contribution of each food group to dietary intakes of EPA, DPA, DHA and total long chain n-3 PUFA was determined by dividing values of EPA, DPA, DHA and total long chain n-3 PUFA of each food group by the total value of each fatty acid for all food groups, and then multiplying by 100 to express it as a percentage contribution.

Statistical analysis

The statistical analysis was carried out using Statistical Package for the Social Sciences (SPSS) software (version 17.0, Chicago IL, USA). Intakes of EPA, DPA and DHA (mg/d) are presented as mean ± SD and median (IQR), adjusted (mg/d/MJ) and unadjusted (mg/d) for energy intake. Intakes of long chain n-3 PUFA were determined by gender, age group (2-3 y, 4-8 y, 9-13 y and 14-16 y) and fish eater status (fish eater or non-fish eater during the two days of the survey), categorized according to whether any fish had been consumed on one or both days of the two-24 hour recalls. The data were found to be extremely skewed using the Shapiro-Wilk test. Since transformation could not produce normal distribution, the average of two days of intake was used to calculate the long chain n-3 PUFA intakes and for further analysis in this study. It has been argued that averaging two days of intake is appropriate when statistical adjustment methods cannot be used to account intra-individual variability, and when usual intake is of interest [26].

The comparisons of n-3 intakes between groups were made using non-parametric tests; the Mann-Whitney U test was used to compare population intakes of EPA, DPA and DHA by gender and fish eater status; the Kruskal-Wallis test was used to compare EPA, DPA and DHA by age group. Correlations between variables were calculated using Kendall tau-b correlation coefficients () as it tends to provide a better estimate of the true population correlation than Spearman’s rho and it is not artificially inflated by multiple tied ranks [27]. A probability of P < 0.05 was taken as statistically significant.

Chapter 2. Omega-3 LCPUFA intake study

Results

The distribution of EPA, DPA, DHA and total long chain n-3 PUFA intake was positively skewed with extreme values in the upper tail; hence values are also provided for median (IQR) intakes. Mean and median intakes of EPA, DPA, DHA and total long chain n-3 PUFA increased with age and tended to be higher in boys than girls, but these trends were not significant (Table 1A). After adjusting for total energy, intakes of EPA (P < 0.05), DPA (P < 0.001) and total long chain n-3 PUFA (P < 0.05) were significantly higher in the older compared to younger age groups; and there were no significant gender differences (Table 1B). The median intakes (mg/1000 Kcal/d) of DPA in non-fish eating children was approximately 50 % of those in fish consumers, whereas the median EPA and DHA intakes were only 12 % and 14 % of the mean intakes (Table 2). The trend of EPA, DPA and DHA intake is the same for all age groups.

The mean consumption of fish over the two days of the survey was 12 g/d, with twenty one percent of all children having consumed fish and/or seafood on at least one of the two days of the survey. Of the fish consumers the mean ± SD fish intake was 59 ± 32 g/d. Fish eaters in the different age groups achieved between 15 % and 22 % of the SDT for long chain n-3 PUFA, while those who did not eat fish during the two days of the survey did not meet the SDT (Table 3). The average intake of EPA, DPA and DHA in fish eaters was 7.5, 2 and 16 fold higher, respectively, than in those who did not eat fish during the two days of the survey (P < 0.001) (Table 3).

EPA ( = 0.46) and DHA ( = 0.52) intakes were highly correlated to consumption of fish and seafood products, and DPA ( = 0.28) was moderately correlated to consumption of meat, poultry and game products (Table 4). Very weak correlations between EPA, DPA and DHA and other food groups were observed (Table 4).

The major food groups contributing to EPA intake were fish and seafood products (59 %) followed by meat, poultry and game products and dishes (24 %). Fish and seafood were the major contributor to DHA intake (75 %), followed by eggs and egg- containing dishes (12 %). The main contributor to DPA was meat, poultry and game products (56 %), followed by fish and seafood (23 %) (Table 5). 64

Chapter 2. Omega-3 LCPUFA intake study

Table 1A. Average daily intake of long chain n-3 PUFA for all children by age group and sex (mg/d) EPA DPA DHA Total LC n-3 PUFAa (mg/d) (mg/d) (mg/d) (mg/d) n Mean ± SD Median Mean ± SD Median Mean ± SD Median Mean ± SD Median (IQR) (IQR) (IQR) (IQR) Age All ages 4,487 21.8 ± 54.7 7.2 18.2 ± 29.5 8.9 39.0 ± 102.4 6.0 79.2 ± 173 28.9 (2.0-20.6) (3.5-20.3) (0.9-26.8) (10.9-72.5) 2-3 y 1,071 16.5 ± 40.5 5.3 12.0 ± 18.6 6.2 31.2 ± 77.2 3.9 59.7 ± 128.5 21.5 (1.5-13.8) (2.2-13.8) (0.6-23.6) (7.3-54.8) 4-8 y 1,216 19.2 ± 47.6 6.7 15.3 ± 23.5 8.2 35.9 ± 92.7 5.1 70.5 ± 152.7 26.1 (1.8-18.0) (3.3-17.6) (0.9-26.5) (9.6-64.2) 9-13 y 1,110 23.5 ± 58.0 8.7 20.8 ± 32.1 10.8 40.9 ± 107.3 6.8 85.3 ± 183.2 32.6 (2.6-23.0) (4.3-23.8) (1.1-27.4) (13.9-75.7) 14-16 y 1,090 28.2 ± 68.9 9.8 24.9 ± 38.9 12.2 49.3 ± 127.0 7.8 102.5 ± 215.9 36.7 (2.7-28.0) (5-28.7) (1.5-32.9) (15.9-94.4) Sex

Boys 2,249 23.2 ± 59.6 7.6 19.3 ± 32.4 9.0 40.9 ± 110.8 6.4 83.5 ± 187.9 29.6 (2-21.9) (3.5-21.5) (0.9-27.5) (11.5-75.5) Girls 2,238 20.4 ± 49.6 6.8 17.1 ± 26.4 8.7 37.6 ± 93.8 5.9 75.0 ± 157.4 27.9 (1.9-18.9) (3.4-19.4) (0.9-26.7) (10.5-69.4)

aTotal LC (long chain) n-3 PUFA: sum of EPA, DPA, and DHA

Chapter 2. Omega-3 LCPUFA intake study

Table 2. Average daily intake of EPA, DPA, DHA and total LC n-3 PUFA for all children by fish eater statusa , adjusted and unadjusted for energy intake Intake (mg/d) Intake (mg/1000 Kcal/d) Fish eater Non-fish eater Fish eater Non-fish eater (n 933) (n 3,554) (n 933) (n 3,554) Mean ± SD Median Mean ± Median Mean ± SD Median Mean ± SD Median (IQR) SD (IQR) (IQR) (IQR) All ages EPA 70 ± 104 36 9 ± 13 5 38 ± 54 21 5 ± 6.3 2.5 (19-72) (1-12) (13-42) (0.8-6.7) DPA 32 ± 48 16 15 ± 21 8 17 ± 21 8.4 7.5 ± 11 4.2 (7-39) (3-17) (4.2-21) (1.7-9.2) DHA 153 ± 182 87 9 ± 16 3 84 ± 96 50 5 ± 8 2.1 (40-203) (0.3-11) (25-109) (0.2-6.3) Total LC 255 ± 314 150 33 ± 41 21 138 ± 163 84 18 ± 21 12 n-3 PUFA (76-308) (8-43) (46-172) (14.6-23)

a Fish eater, ate fish during at least one of the two survey days; Non fish eater, did not eat fish during the two survey days Intake of EPA, DPA, DHA and total LC (long chain) n-3 PUFA (mg/1000 Kcal/d) for fish eater and non-fish eater are significantly different between corresponding values for all age groups (P<0.001) bTotal LC n-3 PUFA: sum of EPA, DPA, and DHA

Chapter 2. Omega-3 LCPUFA intake study

Table 3. The consumption (g/d) of major food sources of long chain n-3 PUFA (meat, egg and fish) by fish eater statusa in relation to percentage of children meeting SDTa

Actual consumption (g/d) Percentage of children c Percentage of children (mean ± SD) meeting the SDT Meat Egg Fish n Non-fish eaters Fish eaters Non-fish Fish eaters Non-fish Fish eater Fish eater Non-fish Fish eaters (%) (%) eater eaters eaters Boys 2-3 y 550 75 25 65 ± 52 47 ± 38 5.3 ± 14 6.5 ± 18 42 ± 35 0.5 16 4-8 y 613 79 21 101 ± 73 78 ± 58 5.8 ± 16 8.8 ± 20 63 ± 51 0.0 16 3-13 y 525 81 19 129 ± 88 110 ± 98 6.0 ± 16 6.7 ± 17 78 ± 66 0.0 15 14-16 y 561 82 18 186 ± 114 138 ± 109 7.5 ± 19 12.6 ± 22 83 ± 58 0.0 21 Girl 2-3 y 521 77 23 68 ± 52 50 ± 40 4.8 ± 12 5.4 ± 11 43 ± 33 0.0 22 4-8 y 603 76 24 84 ± 61 63 ± 58 5.2 ± 13 8.4 ± 16 48 ± 39 0.0 20 3-13 y 585 82 18 113 ± 93 90 ± 73 7.6 ± 20 6.8 ± 18 59 ± 46 0.0 16 14-16 y 529 81 19 112 ± 92 92 ± 84 7.5 ± 20 8.8 ± 23 68 ± 42 0.0 18

Total children 4487 79 21 108 ± 30 80 ±76 6.2 ± 16 7.9 ± 18 59 ± 32 0.0* 18

aFish eater, ate fish during at least one of the two survey days; Non fish eater, did not eat fish during the two survey days bSDT, suggested dietary target cSDT for LC (long chain) n-3 PUFA (Meyer & Kolanu, 2011) [22] total LC n-3 PUFA: sum of EPA, DPA, and DHA *The actual value = 0.000563 n, number of children

Chapter 2. Omega-3 LCPUFA intake study

Table 4. Intakes of EPA, DPA, DHA and total LC n-3 PUFAa in correlation with the intakes of some food groups (Kendall’s tau-b correlation coefficients) EPA DPA DHA Total LC n-3 PUFA EPA 0.525b 0.395 0.692 DPA 0.343 0.633 DHA 0.651 Fish and seafood products and 0.463 0.191 0.519 0.486 dishes Meat, poultry and game products 0.161 0.280 0.114 0.193 and dishes Egg products and dishes 0.023 0.110 0.379 0.242 P < 0.05 Cereal based products and dishes 0.038 0.048 0.023 0.044 P < 0.05 Fats and oils 0.041 0.043 0.015 0.035 Vegetable products and dishes 0.078 0.072 0.029 0.067 P = 0.004 aTotal LC (long chain) n-3 PUFA: sum of EPA, DPA, and DHA; bAll correlation coefficients are significant for P < 0.001, except those for which a P value is indicated

Table 5. Percentage contribution of food groups to EPA, DPA, DHA and total long chain (LC) n-3 PUFAa for all children (n 4,487) aged 2-16 y Food group name EPA DPA DHA Total LC n-3 PUFA Fish and seafood products and dishes 59.4 23.6 75.9 59.3 Meat, poultry and game products and dishes 24.5 56.1 7.1 23.1 Egg products and dishes 0.3 3.6 12.6 7.1 Cereal-based products and dishesb 4.2 5.7 2.2 3.6 Milk products and dishesc 2.3 5.6 0.2 2.0 Fats and oils 1.9 1.9 0.0 1.0 Vegetable products and dishes 2.7 0.4 0.0 0.9 Cereal and cereal products 0.6 0.4 1.2 0.8 Soup 0.7 1.3 0.4 0.7 Snack foods 2.2 0.4 0.0 0.7 Savoury sauces and condiments 0.8 0.5 0.4 0.5 Sugar products and dishes 0.2 0.2 0.0 0.1 Miscellaneous 0.1 0.1 0.0 0.0 Fruit products and dishes 0.1 0.1 0.0 0.0 Confectionery and cereal/nut/fruit/seed bars 0.1 0.1 0.0 0.0 Other: infant formulae and foods, non-alcoholic 0.0 0.0 0.0 0.0 beverages, dairy substitutes, seed and nut products and dishes, legume and pulse products and dishes, alcoholic beverages, special dietary foods aTotal LC (long chain) n-3 PUFA: sum of EPA, DPA, and DHA; bIncluding n-3 enriched bread; c Including n-3 enriched milk 68

Chapter 2. Omega-3 LCPUFA intake study

Discussion

This study quantified the dietary intake of EPA, DPA and DHA in a large, nationally representative sample of Australian children. An extremely low intake of DHA was identified, with half the population in all age groups (median) having extremely low DHA intakes, and this is applicable to all age groups (Table 1). These low DHA intakes can be explained by the fact that on the two days of the survey approximately 80 % of children did not consume fish or seafood, which is the greatest contributor to DHA in the diet (Table 5).

This study confirms that the intakes of EPA, DPA, DHA and total LC n-3 PUFA intakes have a skewed distribution, where very few individuals consume a lot and the vast majority of individuals consume very little LC n-3 PUFA [22]. Even though the distribution of EPA, DPA and DHA intakes are skewed, this study highlights that the Australian median DHA intakes (mg/1000 Kcal/d) is only 13 % of the mean DHA intakes (mg/1000 Kcal/d), compared with the distribution of DPA intakes where the median intakes were 50 % of the mean (refer to Table 1B). These differences can be explained by most children consuming regularly meat (rich source of DPA) and only a small proportion of children consuming regularly fish (rich source of DHA).

DHA is important for brain development up to 18 years of age [28, 29] and even life-long [30]. Since children with amino acid metabolism cannot consume protein rich foods like fish and seafood, they have a significantly lower DHA status in their diets as compared to healthy children [10]. Therefore, it has been suggested that DHA should be considered a semi-essential nutrient [10]. Australian children should therefore increase their intakes of DHA for optimal health [22].

Fish and seafood is the food group that contributes the greatest amount of DHA (Table 5). Our data show that the population average of fish consumption on the two days of the survey is below the national and international recommendations for children, of at least two fish meals per week or approximately 500 mg EPA plus DHA [20, 31], with only one in five children consuming fish. Our current fish intakes were similar to those reported in previous Australian national surveys of children conducted in 1985 (mean = 6 to10 g/d) [32] and 1995 (mean = 12 to 19 g/d) [33]. It can be argued that fish consumption in children can be considered a proxy indicator for assessing the amount of

Chapter 2. Omega-3 LCPUFA intake study

LC n-3 PUFA intake, particularly DHA intake [34]. Frequency of fish consumption in children is significantly correlated with EPA and DHA in the blood serum [35]. Data from the Australian 1995 survey reported that the mean intake of total LC n-3 PUFA in children aged 2-11 and 12-18 years was 110 and 197 mg/d, respectively, of which around 42 % and 38 % for each of the age groups, respectively, were supplied by DHA, with the remainder provided by EPA and DPA [36]. Similar findings have been reported from Belgium, where fish consumption, although low (8.6 g/d), was the main contributor to total EPA, DPA and DHA intakes, providing 53.5, 42.8 and 48.2 %, respectively [37]. A low fish consumption in children has also been reported in other countries, including Germany [38] and Guatemala [39]. A study on 1024 German children of age 2-18 years reported that mean intake of long chain n-3 PUFA (sum EPA + DHA) estimated using yearly 3-day weighed dietary records was 40-140 mg/d. It was lower in those that did not consume fish (below 20 mg/d), but twice as high in those that did eat fish; remaining constant after adjustment for total energy [38]. A study on 449 Guatemalan school children aged 8-12 years reported that mean intake of EPA and DHA estimated using a single pictorial 24-hour record were 9 mg and 32 mg, respectively, higher in boys than in girls, even after adjustment for total energy [39]. Another study on American children aged 6-11 (n = 962) and 12-19 years (n = 2,208) showed that mean (median) intakes of EPA, DPA and DHA obtained using a 24-hour recall were for EPA: 10 (0) and 20 (0) mg/d respectively, DPA: 10 (0) mg/d for both age groups and DHA: 40 (10) and 50 (10) mg/d respectively [40]. An exception is Japan, where the mean fish consumption of children (12-15 years) is 17.9 ± 8.9 g/1000 Kcal, where mean energy intake was 2206 ± 595 Kcal/d, corresponding to an EPA+DHA intake of 0.17 ± 0.1 % of total energy [41]. This level of intake is comparable to the Australian SDT for long chain n-3 PUFA intake in adults, which corresponds to 0.2% of total energy intake [21].

In this study, despite the low fish consumption, fish being a rich source of LC n-3 PUFA contributed 60 % to long chain n-3 PUFA, which is slightly greater than that reported for Australian adults, where 48 % of long chain n-3 PUFA intake was derived from fish and seafood products [36]. In our study, and similarly in Australian adults [36], meat, poultry and game was the second largest food group contributor to long chain n-3 PUFA, but this was primarily due to the contribution of DPA. Two reasons for the high intake of DPA are that Australian children consume at least 8 times more 70

Chapter 2. Omega-3 LCPUFA intake study

meat than fish and that the cattle are grass-fed rather than grain-fed, providing a higher meat DPA content in Australian meat than in other countries [42]. However, the second largest contributor to DHA intake was eggs and not meat. Although the children consumed fewer eggs than meat (Table 2), eggs contain more DHA per gram than meat, which explains their relatively high contribution to DHA intake (Table 5).

Dietary habits in childhood influence the development of an individual's CVD risk profile in later life [43]. Coupled with the increasing prevalence of overweight and obesity in Australian children with increasing age [44, 45], it is timely to consider increasing intake levels of long chain n-3 PUFA as a part of CVD prevention programs for Australian children.

There are several ways to improve the intake of long chain n-3 PUFA, including nutrition education and behavior modification strategies to increase the consumption of fish and seafood, n-3 supplementation, or incorporation of foods that are enriched with n-3, such as certain brands of bread, milk, yoghurt or eggs. Australia is the leading market for n-3 enriched products (e.g. bread enriched with n-3) [46], and Australian consumers have claimed to have awareness of the health benefits of n-3 enriched foods [47]. However, it has been reported that fewer than 7 % of children in the Children’s survey consumed n-3 enriched food products, including bread and milk [22], hence, we did not separate this type of food in our analysis. We have recently reported that long chain n-3 PUFA intakes can be approximately tripled if the usual intakes of bread, milk, eggs and yoghurt are substituted with the same foods that have been enriched with long chain n-3 PUFA [48]. However, these food substitutions did not result in an increase intake sufficiently to meet the SDT for long chain n-3 PUFA in those who did not eat fish during the two days of the survey; the best way remained an increased fish and seafood consumption. With regard to the sustainability of increasing fish consumption, the first national report of the status of key Australian fish stocks in 2012 indicated that at present most fish in Australia are caught in sustainable numbers and that for 88% of stocks (from 111 varieties), the fish biomass was healthy and harvested sustainably. This suggests that our recommendation to increase fish consumption is a viable option, although monitoring is required to ensure future sustainability [49]. Alternatives such as farmed fish may need to become a greater food source for delivery of n-3 LCPUFA. Therefore the challenge is to encourage non-fish eaters to include fish and seafood in

Chapter 2. Omega-3 LCPUFA intake study

their habitual diets, and to overcome reported barriers to intake, including undesirable physical properties (smell and bones), difficulties with preparation and cooking, unaffordability, presence of food allergies, and a perceived risk of pollutants [38, 50- 52]. Communication of health messages that encourage children to include more fish in their diet [53, 54] will be one aspect, while educating parents and caregivers about practical, easy ways to prepare fish dishes may be of higher importance. Additionally, as well as being an excellent source of long chain n-3 PUFA, fish provides several other essential nutrients, including iodine, vitamin D, zinc, magnesium, phosphorus, selenium and potassium [55].

A potential limitation of this study is the use of two 24-hour recalls to obtain estimates of usual long chain n-3 PUFA intake. This may lead to under or over reporting, because fish is consumed only occasionally by most people. Red blood cells fatty acids are a valid biomarker of habitual long chain n-3 PUFA [56], however, the data were not collected in the present dietary survey. A previous national dietary survey in 1995 found that long chain n-3 PUFA intakes quantified from a single 24-h recall was almost identical to that obtained by a FFQ in Australian adults, with a mean and median of 247 and 119 mg/d, respectively [36]. Further, the authors argued that the difference between mean and median intakes, specifically for EPA and DHA, is likely to reflect a low proportion of the population consuming large quantities of fish, rather than low fish consumption by individuals in the 24-h recall [36]. A study using direct quantitative analysis of long chain n-3 PUFA intake in Canadian children aged 4-8 years [57], reported a mean intake long chain n-3 PUFA (EPA plus DHA) that was midway between the values of consumers and non-consumers of fish in the present study. Our data is consistent with previous national dietary surveys conducted in 1985 [32] and 1995 [33], as well as in Western Australia [58], in which the trend of low fish consumption appears to have remained unchanged in the population of Australian children. Similar intakes of fish have also been reported in a recent study of overweight and obese children [59] as well as in a survey of Australian families with young children [47]. It should be noted, however, that the mean and median intakes obtained by 24- hour recall from appropriately sampled group, such as the Children’s Survey can provide a reliable estimate of the average usual intake of the group or population but not the usual intakes of the individual [60].

Chapter 2. Omega-3 LCPUFA intake study

In conclusion, Australian children are not consuming optimal amounts of long chain n-3 PUFA, especially DHA, which could be explained due to a low intake of fish and seafood products.

Acknowledgements

Permission to use the 2007 Australian National Children’s Nutrition and Physical Activity Survey from the Australian Social Science Data Archive is gratefully acknowledged. EPA, DPA and DHA data from the 2007 Australian National Children’s Nutrition and Physical Activity Survey was kindly provided by Judy Cunningham. We would also like to thank the Directorate General Higher Education Indonesia for sponsoring Setyaningrum Rahmawaty, a lecturer from the University of Muhammadiyah Surakarta Indonesia for her PhD at the University of Wollongong, New South Wales, Australia. None of the authors had conflicts of interest to declare.

References

1. Uauy R, Dangour AD (2006) Nutrition in brain development and aging: role of essential fatty acids. Nutr Rev 64:S24-S33

2. Guesnet P, Alessandri JM: Mini-review (2011) Docosahexaenoic acid (DHA) and the developing central nervous system (CNS) – Implication for dietary recommendations. Biochimie 93:7-12

3. Romeo J, Wärnberg J, García-Mármol E, Rodrı´guez-Rodrı´guez M, Diaz LE, Gomez-Martı´nez S et al (2011) Daily consumption of milk enriched with fish oil, oleic acid, minerals and vitamins reduces cell adhesion molecules in healthy children. Nutr Metab Cardiovasc Dis 21:113-120

4. Dangardt F, Osika W, Chen Y, Nilsson U, Gan LM, Gronowitz E et al (2010) Omega-3 fatty acid supplementation improves vascular function and reduces inflammation in obese adolescents. Atherosclerosis 212:580-585

5. Hawthorne KM, Abrams SA, Heird WC (2009) Docosahexaenoic acid (DHA) supplementation of orange juice increases plasma phospholipid DHA content of children. J Am Diet Assoc 109:708-712

6. Lopez-Huertas E (2010) Health effects of oleic acid and long chain omega-3 fatty acids (EPA and DHA) enriched milks: A review of intervention studies. Pharm Res 61:200-207

7. Das UN (2006) Essential fatty acids: biochemistry, physiology and pathology, Biotec J, 1:420-439

Chapter 2. Omega-3 LCPUFA intake study

8. Pawlosky RJ, Hibbeln, Novotny JA, Salem N (2010) Physiological compartmental analysis of α-linolenic acid metabolism in adult humans. J Lipid Res 42:1257-1265

9. Burdge GC, Finnegan YE, Minihane AM, Williams CM, Wootton SA (2003) Effect of altered dietary n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13 C]α-linolenic acid to longer-chain fatty acids and partitioning towards β-oxidation in older men. Br J Nutr 90:311-321

10. Vlaardingerbroek H, Hornstra G, de Koning TJ, Smeitink JAM, Bakker HD, de Klerk HBC et al (2006) Essential polyunsaturated fatty acids in plasma and erythrocytes of children with inborn errors of amino acid metabolism. Mol Genet Metab 88:159-165

11. Russo GL (2009) Dietary n-6 and n-3 polyunsaturated fatty acids: from biochemistry to clinical implication for cardiovascular prevention. Biochem Pharm 77:937-946

12. Riediger ND, Othman RA, Suh M, Moghadasian MH (2009) A Systemic Review of the Roles of n-3 Fatty Acids in Health and Disease. J Am Diet Assoc 109:668-679

13. Ridker PM (2003) Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation 107:363-369

14. Klein-Platat C, Drai J, Oujaa M, Schlienger J, Simon C (2005) Plasma fatty acid composition is associated with the metabolic syndrome and low-grade inflammation in overweight adolescents. Am J Clin Nutr 82:1178-84

15. Harris WS (2008) The omega-3 index as a risk factor for coronary heart disease. Am J Clin Nutr 87(suppl):1997S-2002S

16. Burrows T, Collins CE, Garg ML (2011) Omega-3 index, obesity and insulin resistance in children. Int J Pediatr Obes 6:e532-e539

17. Ross R (1993) The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 362:801-809

18. Roldán V, Marín F, Lip GYH, Blann AD (2003) Soluble E-selectin in cardiovascular disease and its risk factors: a review of the literature. Thromb Haemost 90:1007-1020

19. Engler MM, Engler MB, Malloy M, Chiu E, Besio D, Paul S et al (2004) Docosahexaenoic acid restores endothelial function in children with hyperlipidemia: results from the EARLY Study. Int J Clin Pharmacol Ther 42:672– 679

20. National Heart Foundation of Australia (2008) Position statement, fish, fish oils, n-3 polyunsaturated fatty acids and cardiovascular health. Available at: http://www.heartfoundation.org.au/sitecollectiondocuments/fish-position- statement.pdf: Accessed January, 15, 2011

Chapter 2. Omega-3 LCPUFA intake study

21. National Health Medical Research Council (2006) Nutrient reference value for Australia and New Zealand including recommended dietary intakes. Australian Government Department of Health and Ageing. Commonwealth of Australia. Available at: http://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/n35.pdf: Accessed June, 6, 2010

22. Meyer BJ, Kolanu N (2011) Australian children are not consuming enough long- chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 27:1136- 1140

23. Commonwealth of Australia (2008) Main findings - 2007 Australian National Children’s Nutrition and Physical Activity Survey. Commonwealth of Australia

24. Australia Social Science Data Archive (2009) Available at: http://assdanesstar .anu.edu.au/webview/?object=http://assda-nesstar.anu.edu.au/obj/fCatalog/Catalog 28. Accessed January, 7, 2009

25. Department of Health and Ageing (2007) User guide - 2007 Australian national children’s nutrition and physical activity survey. Department of Health and Ageing. Available at: http://www.health.gov.au/internet/main/publishing.nsf/Content/phd- nutrition-childrens-survey-userguide. Accessed June, 23, 2010

26. Department of Health and Ageing (2012) The 2007 Australian National Children’s Nutrition and Physical Activity Survey Volume Eight: Dietary exposure to food secured chemicals. Department of Health and Ageing. Canberra

27. Allen P, Bennett K (2008) SPSS for the health and behavioural sciences. Thomson, Australia

28. Lauritzen L, Hansen HS, Jùrgensen MH, Michaelsen KF (2001) The essentiality of long chain n-3 fatty acids in relation to development and function of the brain and retina. Prog Lipid Res 40:1-94

29. Carver JD, Benford VJ, Han B, Cantor AB (2001) The relationship between age and the fatty acid composition of cerebral cortex and erythrocytes in human subjects. Brain Res Bull 56:79-85

30. Parletta N, Milte C, Meyer BJ (2013) Nutritional modulation of cognitive function and mental health. J Nutr Biochem 24:725-743

31. Kris-Etherton PM, Grieger JA, Etherton TD (2009) Dietary reference intakes for DHA and EPA. Prostaglandins Leukot Essent Fatty Acids 81:99-104

32. English R, Cashel KM, Lewis JL, Bennet S, Berzins J, Penm R (1988) National Dietary Survey of Schoolchildren aged 10–15 Years. Report no. 1. Foods Consumed. Canberra: Australian Government Publishing Service

Chapter 2. Omega-3 LCPUFA intake study

33. McLennan W, Podger A (1999) National Nutrition Survey – Foods Eaten, Australia. Canberra: Australian Bureau of Statistics, Commonwealth Department of Health and Aged Care

34. Amiano P, Dorronsoro M, de Renobales M, Ruiz de Gordoa JC, Irigoien I, EPIC Group of Spain (2001) Very-long-chain ω-3 fatty acids as markers for habitual fish intake in a population consuming mainly lean fish: the EPIC cohort of Gipuzkoa. Eur J Clin Nutr 55:827-832

35. Oddy WH, Sherriff JL, Kendall GE, De Klerk NH, Mori TA, Blake KV et al (2004) Patterns of fish consumption and levels of serum phospholipid very-long-chain omega-3 fatty acids in children with and without asthma, living in Perth, Western Australia. Nutr Diet 61:30-37

36. Howe P, Meyer B, Record, Baghurst K (2006) Dietary intake of long-chain ω-3 polyunsaturated fatty acids: contribution of meat sources. Nutrition 22:47-53

37. Sioen I, Huybrechts I, Verbeke W, Van Camp J, De Henauw S (2007) n-6 and n-3 PUFA intakes of pre-school children in Flanders, Belgium. Br J Nutr 98:819-825

38. Sichert-Hellert W, Wicher M, Kersting M (2009) Age and time trends in fish consumption pattern of children and adolescents, and consequences for the intake of long-chain n-3 polyunsaturated fatty acids. Eur J Clin Nutr 63:1071-1075

39. Bermudez OI, Toher C, Montenegro-Bethancourt G, Vossenaar M, Mathias P, Doak C et al (2010) Dietary intakes and food sources of fat and fatty acids in Guatemalan school children: A cross-sectional study. J Nutr 9:20

40. Ervin B, Wright JD, Wang CY, Kennedy-Stephenson J, Division of Health and Nutrition Examination Surveys (2004) Dietary Intake of fats and fatty acids for the United States population: 1999–2000. Advanced data from vital health statistic, CDC Number 348: November 8

41. Murakami K, Miyake Y, Sasaki S, Tanaka K, Arakawa M (2010) Fish and n-3 Polyunsaturated Fatty Acid Intake and Depressive Symptoms: Ryukyus Child Health Study. Pediatrics 126:e623-e630

42. Droulez V, William PG, Levy G, Stobaus T, Sinclair A (2006) Composition of Australian red meat 2002. 2. Fatty acid profile. Food Aust 58:335-341

43. Niinikoski H, Lagström H, Jokinen E, Siltala M, Rönnemaa T, Viikari J et al (2007) Impact of repeated dietary counselling between infancy and 14 years of age on dietary intakes and serum lipids and lipoproteins: the STRIP study. Circulation 116:1032-1040

44. Venn AJ, Thomson RJ, Schmidt MD, Cleland VJ, Curry BA, Gennat HC et al (2007) Overwight and obesity from childhood to adulthood: a follow-up of participants in the 1985 Australian School Health and Fitness Survey. Med J Aust 186:458-460

Chapter 2. Omega-3 LCPUFA intake study

45. Cancer Council Australia, Heart Foundation (2011) High school students graduating to be tomorrow’s cancer and cardiovascular patients. Available at: http://www.heartfoundation.org.au/SiteCollectionDocuments/National-student- survey.pdf: Accessed June, 11, 2011

46. Zak L (2008) Of Australia’s manufacturers. Food Mag, June: 16

47. Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Nutrition and Dietetics 2013 (DOI: 10.1111/1747- 0080.12022)

48. Rahmawaty S, Lyons-Wall P, Charlton K, Meyer BJ (2013) Effect of replacement of bread, egg, milk and yogurt with n-3 enriched for these foods on n-3 LCPUFA intake of Australian children. Br J Nutr – resubmitted after corrections for publication

49 Flood, M, Stobutzki, I, Andrews, J, Begg, G, Fletcher, W, Gardner, et al (eds) (2012) Status of key Australian fish stocks reports 2012, Fisheries Research and Development Corporation, Canberra. Available at: http://fish.gov.au/Pages /SAFS_Report.aspx

50. Leek S, Muddock S, Foxall G (2000) Situational determinants of fish consumption. Br Food J 102:18-39

51. Verbeke W, Vackier I (2005) Individual determinants of fish consumption: application of the theory of planned behaviour. Appetite 44:67-82

52. Trondsen T, Scholderer J, Lund E, Eggen AE (2003) Perceived barriers to consumption of fish among Norwegian women. Appetite 41:301-314

53. Mozaffarian D, Rimm EB (2006) Fish intake, contaminants, and human health evaluating the risks and the benefits. J Am Med Assoc 296:1885-1899

54. Food and Agriculture Organization/World Health Organization (2011) Report of the joint FAO/WHO expert consultation on the risk and benefits of fish consumption, 25-29 January 2010, Rome, Italy. FAO Fisheries and Aquaculture Report, No 978

55. Ruxton CHS (2011) Review the benefits of fish consumption. Nutr Bull 36:6-19

56. Kuratko CN, Salem N (2009) Biomarkers of DHA status. Prostaglandins Leukot Essent Fatty Acids 81:111-118

57. Madden SMM, Garrioch CF, Holub BJ (2009) Direct diet quantitation indicates low intakes of (n-3) fatty acids in children 4 to 8 years old. J Nutr 139:528-532

Chapter 2. Omega-3 LCPUFA intake study

58. Clayton EH, Hanstock TL, Watson JF (2009) Estimated intakes of meat and fish by children and adolescents in Australia and comparison with recommendations. Br J Nutr 101:1731-1735

59. Burrows T, Berthon B, Garg ML, Collins CE (2012) A comparative validation of a child food frequency questionnaire using red blood cell membrane fatty acids. Eur J Clin Nutr 66:852-829

60. Rutishauser I (2000) Getting it right – how to use data from the 1995 National Nutrition Survey. Commonwealth of Australia, Canberra

Chapter 3. Fish and seafood survey

Chapter 3

The majority of this section is the substantive content of the work published in Nutrition and Dietetics 2013, DOI: 10.1111/1747-0080.12022

The findings of this study were also presented in the 36th Nutrition Society of Australia, Wollongong 2012 (poster presentation). Abstract was published in the Proceedings of the Nutrition Society of Australia 36, 70.

Statement

As the primary supervisor, I, Barbara J Meyer, declare that the greater part of the work in this paper is attributed to the candidate, Setyaningrum Rahmawaty.

In the manuscript, Setyaningrum contributed to study design and was primarily responsible for data collection, data analysis, data interpretatio and writing up the manuscript.

A/Prof Barbara J Meyer (Main Supervisor)

Chapter 3. Fish and seafood survey

Factors that influence consumption of fish and omega-3 enriched foods: A survey of Australian families with young children

(Published in journal Nutrition and Dietetics 2013, DOI: 10.1111/1747-0080.12022)

Setyaningrum RAHMAWATY1,2, Karen CHARLTON2, Philippa LYONS- WALL3, Barbara J MEYER1,2

1Metabolic Research Centre, University of Wollongong: 2School of Health Sciences, University of Wollongong, Wollongong, New South Wales, and 3School of Exercise and Health Sciences, Edith Cowan University, Perth, Western Australia, Australia

Abstract

Aim: The present study aimed to identify factors that influence the consumption of fish and foods that are enriched with omega-3 long chain polyunsaturated fatty acids (n-3 LCPUFA), in order to inform the development of effective nutrition education strategies.

Methods: A cross-sectional, ten-item self-administered survey was conducted in 262 parents of children aged 9-13 years from a regional centre in Wollongong. Parents were asked questions related to frequency of consumption, and to identify factors that either encouraged or prevented the provision of fish/seafood and/or n-3 LCPUFA enriched foods for their families.

Results: Salmon, canned tuna, prawn and take-away fish were the most commonly eaten variants of fish/seafood, at approximately once a month. Perceived health benefits, and the influence of media and health professionals in health promotion were identified as the primary motivators for consumption of fish/seafood and foods enriched with n-3 LCPUFA. Among families who consume fish, taste was valued as having a major positive influence, as well as preferences of individual family members, but the latter was perceived as an obstacle in non-fish consumers. Price was the main barrier to consumption of fresh, but not canned, fish and n-3 enriched foods, in both those that do and do not consume these foods.

Chapter 3. Fish and seafood survey

Conclusion: Despite Australian parents’ knowledge of the health benefits n-3 LCPUFA, only a fifth of households meet the recommended two serves of fish per week, hence nutrition education strategies are warranted.

Keywords: barriers, beliefs, fish consumption, omega-3 enriched food, parents, promoting factors. ______

S. Rahmawaty, MHSc, PhD student K. Charlton, PhD APD, A/Professor P. Lyons-Wall, PhD APD, A/Professor B.J. Meyer, PhD, RNutr, A/Professor Correspondence: B.J. Meyer, School of Health Sciences, University of Wollongong, Northfields Ave, Wollongong NSW 2522, Australia. Email: [email protected]

Introduction

The health benefits of consuming fish and/or seafood and foods enriched with n-3 have been widely reported in school children and adolescents, with outcomes including improvements in cognitive performance and academic achievement,1-5 atheroprotective lipid profiles,6 improved cardiovascular risk factors7 and possibly the prevention of obesity- related chronic diseases.8 These benefits are primary attributed to the high content of n-3 LCPUFA, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). These fatty acids are insufficiently de novo synthesized in the human body,9,10 thus, must be obtained from foods containing preformed EPA and DHA, such as fish and/or seafood11,12 as well as foods enriched with n-3.13 Fish consumption is a valid predictor for EPA and DHA concentration in serum phospholipid.14,15 Likewise, the consumption of omega-3 enriched foods (e.g. bread, egg, milk and dips) leads to a significantly increased concentration of EPA and DHA levels in serum phospholipids16,17 plasma18,19 and erythrocytes.20

The Australian Dietary Guidelines advise children to consume a half to one serving of fish each day, with a serve size equivalent to 80 - 120 grams of cooked fish fillet for optimal brain development and cardiovascular health.21 Other agencies such as the National

Chapter 3. Fish and seafood survey

Heart Foundation of Australia22,23 recommend higher intakes (2 or 3 serves of 150 g per serve size of fatty fish per week), as well as food and drinks enriched with n-3 LCPUFA.

Fish and/or seafood products have been reported to provide 88% of DHA, 78% of EPA and 79% of total n-3 LCPUFA in the diets of Australian children.12 Consumers of fish ate on average 59 g (SD 48 g) per day, which enabled them to meet approximately 70% of the suggested dietary target (SDT) for n-3 LCPUFA.12 However, in a nationally representative sample of children, only a fifth had eaten fish within the two days of dietary reporting.24,25 Additionally, less than 7% of the Australian children consumed foods enriched with n-3,25 which has previously been proposed as a strategy to increase population-intake of n-3 LCPUFA.26,27 In countries with traditionally low fish consumption such as Australia,25 n-3 enriched foods may play an important role in meeting the dietary recommendation for n-3 LCUFA intake for optimal health. Such foods are available in supermarkets in Australia, and include various n-3 enriched breads, eggs, milk and yoghurts.

Food consumption patterns of children are largely influenced by individual food preferences28 which in turn may be influenced by parental influences.29 Children are known to model behaviour related to food selection on the dietary habits of their parents.30,31 For example, children’s fish consumption patterns are similar to those of their mother.32

The development of food-acceptance patterns of a child that tend to persist into adulthood is dependent to a large extent on the home environment.33 Involvement of adolescents (9-14 years old) at the family dinner table improved the quality of their diet.34 It is suggested that interventions targeting healthy eating behaviours in children should involve family members in order to modify home environments, whereby parents and other family members in the household also change dietary behaviours to include healthier options30,35 that in the case of increased n-3 intake would include fish and n-3 enriched foods. Fish and/or seafood consumption is influenced by many factors. Numerous identified barriers in adults include difficulty purchasing, preparing and cooking fish; the high cost; the unpleasant physical properties of some characteristics of fish (e.g. bones and smell); a low awareness of the health benefits associated with fish consumption; a limited

Chapter 3. Fish and seafood survey

supply and quality of fresh fish; preferences of other family members; the lack of product choice; unpleasant taste; low socioeconomic background; as well as environmental concerns related to sustainability and the presence of pollutants in the food supply.36-38 A negative attitude towards both the smell and the accompaniments, and fear of finding bones has been reported as barriers for fish consumption in adolescents.39 Food neophobia, ‘avoidance of and reluctance to taste of unfamiliar food’40 has also been identified as a factor that predicts low fish consumption in young adults.41

A number of consumer surveys have investigated fish consumption patterns in groups across Australia,42-45 while others have identified consumers’ intentions to purchase foods enriched with n-3.46,47 The current study is the first one to identify factors which influence parents or primary caregivers decisions to provide fish/seafood and/or n-3 enriched foods to their family. Findings of this study will inform the development of more targeted health messages to improve n-3 LCPUFA intake in Australian children.

Methods

A cross sectional survey was conducted from June to September 2011 in Wollongong, New South Wales, to identify factors relating to parents or primary caregivers’ behaviours in preparing fish and/or seafood and n-3 enriched food products for their family. Two methods of questionnaire administration were offered, either a paper-based survey or an online format using the Survey Monkey tool. A survey questionnaire was developed based on published literature related to barriers and promoting factors affecting consumption of fish and n-3 enriched foods. Ten closed questions addressed about frequency of consumption of various types of fresh or frozen and canned fish/seafood and n-3 enriched food products (three items), barriers and promoting factors for consuming these foods (three items), and preferences for nutrition education activities aimed at improving children’s n-3 LCPUFA intake (four items). An 8-point Likert scale was used for participants’ responses to frequency of consumption (almost never, once a month, 2 or 3 times a month, once a week, 3 or 4 times a week, once a day and not applicable), and a 5- point scale for listed factors that influence consumption patterns (have a large positive influence, some positive influence, neutral, some negative influence and have a large

Chapter 3. Fish and seafood survey

negative influence). The questionnaire was pilot-tested in 7 parents (data not shown) prior to the main survey.

A convenient sample of 262 participants (102 and 160 filled out online and paper- based survey respectively) was recruited using advertisements at the University of Wollongong, child minding centres, shopping centres, mother groups, public offices and local media units in Wollongong. Respondents who were eligible to participate in this survey were parents or primary caregivers who have a child aged 9 to 13 years.

Data was analysed according to categorization into three patterns of fish consumption: frequent eaters (respondents that consumed fish and/or seafood as per the recommendation of 2 or more servings per week), occasional eaters (fish and/or seafood less than twice per week) and non-fish eaters. Participants’ responses related to barriers and promoting factors for fish and/or seafood and n-3 enriched foods were reported in two categories, positive and negative factors (response to neutral was not shown). This study was approved by the the Human Research Ethics Committee at the University of Wollongong.

Results

Of the 262 participants (102 and 160 filled out online and paper-based survey respectively), three-quarters were Australian (76%) with the remainder from various nationalities and most were between 35-44 years of age (62%). Almost all (90%) parents or primary caregivers reported being consumers of fish/ seafood and 20% of households meet the recommended two serves of fish per week. Of those who ate fish, approximately three- quarters consumed it at least once a week, and a third consumed it twice or more per week (Table 1).

Certain types of fresh or frozen fish, including mackerel and herring (77%), anchovies and sardines (76%), and oyster (66%) were almost never eaten (Table 2). Similarly, variants of canned fish or seafood, such as herring and shellfish (80%), mackerel (79%), prawn and fish paste (78%), oysters (76%), anchovies (74%), and sardines (70%) were infrequently consumed (Table 3).

Chapter 3. Fish and seafood survey

Table 1. Characteristic of participants (n =262)

n % Nationality (country of birth used)* Australia 200 76.3 English 11 4.2 Chinese 5 1.9 European 5 1.9 Indonesian 4 1.5 Italian 3 1.1 Indian 3 1.1 Korean 3 1.1 Thai 3 1.1 Lebanese 2 0.8 Turkish 2 0.8 Sri Lankan 2 0.8 Canadian 2 0.8 Irish 1 0.4 Scottish 1 0.4 German 1 0.4 Vietnamese 1 0.4 American 1 0.4 Argentinean 1 0.4 Greek 1 0.4 Macedonian 1 0.4 Midle Costean 1 0.4 New Zealander 1 0.4 Persian 1 0.4 Portuguese 1 0.4 Saudi Arabia 1 0.4 Spanish 1 0.4 Swedish 1 0.4 Welsh 1 0.4 Age (years) <35 52 20 35-44 162 62 45-54 48 18 Fish eater status Frequent eaters (consume ≥ 2 meals fish per week) 54 21 Occasional eaters (consume < 2 meals fish per week) 181 69 Non-fish eater 27 10 * Total % of participant below 100%, because they did not answer the questions

Chapter 3. Fish and seafood survey

The most frequently consumed types of fish were fresh tuna, prawns and salmon, while canned fish favourites included tuna and salmon (Table 2 and 3). In the occasional consumers, fresh prawns (64%) and take-away fried fish and chips (60%) were the most commonly eaten, followed by tuna (34%) and shellfish (26%), whilst in this group, canned fish choices included tuna (59%) and salmon (41%), followed by sardines (14%) and anchovies (11%) (Table 2 and 3).

Table 2. Frequency for consuming fresh or frozen fish or seafood products (% of total participants, n = 262)

Almost Once 2 or 3 Once Twice 3 or 4 Once Never never a month times a week a week times a day a month a week Anchovies* 76 5 1 1 0 0 0 16 Mackerel* 77 3 1 0.4 0 0 0 18 Herring* 77 3 2 0.4 0 0 0 18 Salmon* 34 27 18 8 1 0.4 0.4 10 Tuna* 45 16 12 6 2 3 1 14 Sardines* 76 3 2 0.4 0.4 0.4 0 18 Shellfish* 59 18 6 2 0.4 0 0 14 Oyster* 66 15 3 0.4 0 0.4 0 14 Prawns* 2 8 42 17 5 2 0 0 6 Fish from fish 34 40 15 5 0.4 0 0.4 5 & chips shop* * Total % of participant below 100%, because they did not answer the questions

Table 3. Frequency for consuming canned fish or seafood products (% of total participants, n = 262) Almost Once 2 or 3 Once Twice a 3 or 4 Once Never never a month times a week week times a day a month a week Anchovies 74 8 4 0.4 0 0 0 14 Mackerel 79 2 0.8 0.4 0 0 0 18 Herring 80 2 0.8 0 0 0 0 17 Salmon 47 25 11 6 3 0.4 0 8 Tuna 17 22 21 16 9 8 0.8 5 Sardines* 70 9 4 2 1 0 0 14 Shellfish 80 2 0.4 0 0.4 0 0 17 Oyster 76 6 2 0 0.8 0 0 16 Prawns 78 5 2 0.4 0 0 0 16 Fish paste 78 4 1 0.4 0.4 0 0 16 * Total % of participant below 100%, because they did not answer the questions 86

Chapter 3. Fish and seafood survey

Figure 1 shows the barriers and promoting factors for consuming fresh or frozen and canned fish/seafood according to the three categories of consumption. In all groups, health benefit was the most commonly reported influence (37% to 80%), followed by media (20% to 39%), health promotion activities (15% to 36%), and advice from a health professional (19% to 33%). Identified barriers included dislike for bones (24% to 58%), unpleasant smell (30% to 40%) and pollutant content (22% to 39%). Of the respondents who were consumers of fish, taste was the second most important influence (33% to 57%); in contrast it was a barrier in individuals who were non-consumers (19% to 22%). Almost half of frequent consumers were influenced by preferences of other family members (fresh or frozen fish (44%), canned fish (37%)), while this was rated as a barrier in respondents who were occasional (28%) or non-consumers (22%). Price and difficulties in preparation were also considered as obstacles in the occasional (36 and 28%, respectively) and non- consumer groups (19%). About a third in each category identified the influence of health professionals, health promotion activities and the mass media as being promoting factors for fish consumption. Only about a tenth was influenced by their social group to consume fish.

Chapter 3. Fish and seafood survey

Frequent eater (n = 54) Occasional eater (n = 181) Non-fish eater (n = 27)

Positive Negative Positive Negative Negative Positive

Health benefit 80 1 Smell 33 4

Health benefit 74 2 Taste 52 23 Bone 33 11

Taste 57 13 Health promotion 36 2 Poisoning or pollutant 26 4

Family member 44 19 Media 35 2 Family member 22 11 Media 39 4 Health professional 29 2 Taste 19 15 Price 37 37 Cooking 27 22 Preparing 19 11 Cooking 33 19 Family member 24 28 Price 19 15 Supply in the market 31 31 Preparing 23 28 Variation 11 15 Preparing 31 19 Variation 23 22 Supply in the market 11 4 Health professional 31 4 Supply in the market 23 21 Cooking 7 11 Variation 30 33 Price 19 36 Friends or colleagues 7 11 Health promotion 30 4 Friends or colleagues 14 4 Social group 4 Friends or colleagues 22 2 Social group 12 3 Allergy 4 Smell 15 30

Social group 15 4 Smell 9 40 Health benefit 44

Poisoning or pollutant 11 33 Poisoning or pollutant 7 25 Health promotion 26

Bone 7 52 Bone 6 58 Media 22

Allergy 4 9 Allergy 4 4 Health professional 19

Fig. 1A. Factors that encourage (positive) and prevent (negative) for consuming fresh/frozen fish/seafood (% of respondents) Frequent eater, consume ≥ 2 meals fish per week; Occasional eater, consume < 2 meals fish per week; Non-fish eater, did not eat fish

Chapter 3. Fish and seafood survey

Frequent eater (n = 54) Occasional eater (n = 181) Non-fish eater (n = 27)

Positive Negative Positive Negative Negative Positive

Health benefit 57 9 Health benefit 56 3 Bone 33 7

Taste 52 17 Taste 33 29 Smell 30 4

Supply in the market 41 15 Supply in the market 32 4 Taste 22 15

Price 41 24 Price 32 15 Poisoning or pollutant 22 4 Religion 15 Cooking 41 7 Preparing 30 9 Supply in the market 7 22 Preparing 37 7 Cooking 27 7 Price 7 19 Family member 37 9 Media 28 Preparing 7 19 Media 37 7 Health promotion 28 Cooking 7 11 Variation 35 11 Family member 22 18 Family member 7 7 Health professional 33 2 Health professional 22 Variation 7 19 Health promotion 26 7 Variation 18 10 Health benefit 37 Friends or colleagues 20 2 Friends or colleagues 9 1 Health promotion 15 Social group 17 6 Social group 8 1 Media 20

Bone 15 24 Smell 7 39 Health professional 19

Smell 15 31 Bone 7 32 Friends or colleagues 4

Poisoning or pollutant 4 39 Poisoning or pollutant 7 23 Social group 7

Allergy 2 9 Allergy 3 4 Allergy 4

Fig. 1B. Factors that encourage (positive) and prevent (negative) for consuming canned fish/seafood (% of respondents) Frequent eater, consume ≥ 2 meals fish per week; Occasional eater, consume < 2 meals fish per week; Non-fish eater, did not eat fish

Chapter 3. Fish and seafood survey

Approximately a third of participants almost never consumed foods enriched with n-3, such as bread, egg, yogurt, milk, spread and cereal (Figure 2). Of the participants that did consume n-3 enriched foods, milk, bread and spreads were the highest (Figure 2). Health benefit (47%) was rated as being the most important factor influencing consumption of n-3 enriched food, followed by media (26%), health promotion (25%), health professional advice (23%) and the influence of family members (18%), while price (21%) was rated as the main barrier (Figure 3).

45 40

35 Almost never 30 Once a month 25 2 or 3 times a month 20 Once a week 15 Twice a week % of respondents of % 10 3 or 4 times a week 5 Once a day 0 n-3 enriched n-3 enriched n-3 enriched n-3 enriched n-3 enriched n-3 enriched bread egg yogurt milk spread cereals

Fig. 2. Frequency for consuming omega-3 enriched foods (% of total respondents, n = 262)

Positive Negative

Price 15 21 Supply in the market 18 10 Taste 16 9 Smell 8 8 Variation 15 8 Dose 10 4 Family member 18 3 Media 26 2 Health promotion 25 Friends or colleagues 9 Social group 9 Health benefit 47 Health professional 23

Fig. 3. Factors that encourage (positive) and prevent (negative) for consuming omega-3 enriched foods (% of total respondents, n = 262) 90

Chapter 3. Fish and seafood survey

Regarding preferred nutrition education strategies to improve their children’s n-3 LCPUFA intake, participants identified cooking courses (35%) and group discussions (18%) and they also preferred types of accompanying materials such as recipe books (51%), leaflets/brochures (47%) and online programme materials (23%).

Discussion

The present survey has identified factors that influence decision-making of parents regarding provision of fish and seafood, and foods enriched with n-3 LCPUFA, to their children. Despite a high level of awareness regarding the health benefits of fish identified in the current study and others,48,49 only a fifth of participants reported eating fish and/or seafood at least twice per week, which is the recommendation for cardiovascular health.22,23 Intake of commercially available foods enriched with n-3 was similarly low. Nationally representative data has reported that Australian children and adolescents have low intakes of fish,24,25 which results in inadequate intakes of n-3 LCPUFA.25 A similar finding has been reported in younger pre-school children aged 1 – 5 years, in which only 32 % met the adequate intake for n-3 LCPUFA50 as well as in overweight and obese children aged 5 – 12 years.51

Children have a strong influence on the type of foods prepared and consumed by families. It has been reported that young children’s dislike of fish and seafood may lead to omission of this food group in the family meal repertoire,37,52 while the presence of adolescents in the home is more likely to result in the presence of negative perceptions associated with seafood, including smell during preparation, taste and reported overall family dislike of these foods.52 Further, larger households are associated with a greater dislike of fish and seafood, particularly if those households include children older than 8 years.37 It has been reported that parental involvement has some promising positive impacts in encouraging healthy eating practices in children30,31,34,35,53,54 as well as introducing novel foods enriched with EPA and DHA.27

As well as frequency of fish consumption, we were interested in specific choices thereof, as the n-3 LCPUFA content differs substantially between species. Our data found that most families do not consume EPA and DHA-rich sources of fish and seafood,55 such 91

Chapter 3. Fish and seafood survey

as mackerel, herring, anchovies, sardines and oyster. Except for tuna (mostly canned) and salmon, oily fish was infrequently consumed. Nearly half of the sample reported consuming take-away fish and chips at least once a month, which is twice the frequency reported in 1999 (27%).43 It is noteworthy that the method of cooking does not necessarily affect the amount of EPA and DHA in the fish, however, if the fish is fried using an n-6 oil (e.g. sunflower or safflower), the fish absorbs the n-6 oil, such that the ratio of n-3 to n-6 decreases and this has been shown in fried salmon,56,57 codfish, hake, sole58 mackerel and sardines.57 Thus, the promotion of oily fish to improve n-3 LC PUFA intakes may require attention to cooking method thereof.

It is useful to analyse factors that affect consumption according to frequency of fish intake, in order to identify determinants of behaviour that may be targeted in intervention strategies targeted at families. For example, in those participants that frequently ate fish, the influence of a family member was rated as a driving factor, whereas this was perceived as an obstacle to consumption in the occasional or non-fish eating groups. It has been argued that decisions to consume fish are shaped by three influences, including “favourable attitude, high subjective norm and high perceived behavioural control”,37 while, self- efficacy (confidence to consume) is a necessary predictive factor to consumption of novel sources of n-3 LCPUFA containing foods59,60 as well as the believe that behavioural change will reduce a health risk.59

Our findings regarding barriers to fish consumption are similar to other studies37- 39,61 and include unpleasant physical properties of some varieties of fish (e.g. bones and smell), environmental concerns such as presence of pollutants, difficulties associated with preparation of fish, and the high cost. Price was identified as the most predominant barrier to consuming foods enriched with n-3 LCPUFA, but not for canned fish. In non-fish eaters, taste and allergy were important reasons, as has been reported by others,62 while cost and possibility of overdosing (i.e. erroneous assumption that you could consume too much n-3 LCPUFA from n-3 LCPUFA enriched foods) were the major barriers to eating foods enriched with n-3 LCPUFA.46 Australian consumers argue that external economic policy influences to reduce the price of fish commodities, and food policy that results in more informative labelling on foods (related to “free from contaminants”, such as mercury) are 92

Chapter 3. Fish and seafood survey

required to empower consumers to change dietary practices associated with fish and seafood consumption.46,47 Similarly, clear labelling that convince consumers of the legislation controlling claims as well as the possibility of overdosing on n-3 enriched food package is required to motivate increased consumption thereof.46 Parents in the present study identified that improved cooking skills would be beneficial to encourage their family to eat more fish and n-3 enriched foods.

The main limitation of the study design relates to the use of a convenience sample, drawn from a single regional geographical location. However, external validity can be assessed by comparing the reported fish consumption habits with data from a nationally representative survey of 1025 Australian adults conducted by the Fisheries Research and Development Corporation (FRDC) in 2011,44 in which the same proportion of participants reported consuming fish more than once per week, namely a fifth.

Conclusions

Australian parents are knowledgeable about the health benefits of foods that contain n-3, especially fish, and are also aware of fish being a potential source of pollutants e.g. mercury. Strategies that attempt to alter dietary behaviour in children and adolescents need to focus on the identified barriers to consumption of n-3-rich foods, including perceived family dislike of taste, inconvenience associated with preparation (smell, bones), lack of cooking skills, and cost constraints. The high intake of canned fish variants may necessitate promotion of these alternatives within recipes or as snack foods by nutrition professionals.

Acknowledgements

We thank to the participants participated in this present study. We also thank to the Directorate General Higher Education Indonesia for sponsoring Setyaningrum Rahmawaty, a lecturer from the University of Muhammadiyah Surakarta Indonesia for her PhD at the University of Wollongong, New South Wales, Australia.

Competing interests None identified.

Chapter 3. Fish and seafood survey

REFERENCES

1 Murakami K, Miyake Y, Sasaki S, Tanaka K, Arakawa M. Fish and n-3 Polyunsaturated Fatty Acid Intake and Depressive Symptoms: Ryukyus Child Health Study. Pediatrics 2010; 126: e623-e630.

2 Kim JL, Winkvist A, Åberg MA et al. Fish consumption and school grades in Swedish adolescents, a study of the large general population. Acta Paediatr 2010; 99: 72-77.

3 Åberg MA, Åberg N, Brisman J, Sundberg R, Winkvist A, Torén K. Fish intake of Swedish male adolescents is a predictor of cognitive performance. Acta Paediatr 2009; 98: 555-560.

4 Dalton A, Wolmarans P, Witthuhn RC, Stuijvenberg ME, Swanevelder SA, Smuts CM. A randomised control trial in schoolchildren showed improvement in cognitive function after consuming a bread spread, containing fish flour from a marine source. Prostaglandins Leukot Essent Fatty Acids 2009; 80: 143-149.

5 Groot RHM, Ouwehand C, Jolles J. Eating the right amount of fish: Inverted U- shape association between fish consumption and cognitive performance and academic achievement in Dutch adolescents. Prostaglandins Leukot Essent Fatty Acids 2012; 86: 113-7.

6 Gump BB, MacKenzie JA, Dumas AK et al. Fish consumption, low-level mercury, lipids, and inflammatory markers in children. Environ Res 2012; 112: 204-211.

7 Romeo J, Wa¨rnberg J, Garcı´a-Ma´rmol E et al. Daily consumption of milk enriched with fish oil, oleic acid, minerals and vitamins reduces cell adhesion molecules in healthy children. Nutr Metab Cardiovasc Dis 2011; 21: 113-120.

8 Burrows T, Collins CE, Garg ML. Omega-3 index, obesity and insulin resistance in children. Int J Pediatr Obes 2011; 6: e532-e539.

9 Pawlosky RJ, Hibbeln, Novotny JA, Salem N. Physiological compartmental analysis of α-linolenic acid metabolism in adult humans. J Lipid Res 2001; 42: 1257-1265.

10 Burdge GC, Finnegan YE, Minihane AM, Williams CM, Wootton SA. Effect of altered dietary n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13 C]α-linolenic acid to longer-chain fatty acids and partitioning towards β-oxidation in older men. Br J Nutr 2003; 90: 311-321.

11 Meyer BJ, Mann NJ, Lewis JL, Milligan GC, Sinclair AJ, Howe PRC. Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids. Lipids 2003; 38: 391-398.

12 Rahmawaty S. Dietary intake and food sources of omega-3 long chain EPA, DPA and DHA of Australian children. University of Wollongong, PhD Thesis 2012 (unpublished)

Chapter 3. Fish and seafood survey

13 Lovegrove JA, Brooks CN, Murphy MC, Gould BJ, Williams CM. Use of manufactured foods enriched with fish oils as a means of increasing long-chain n-3 polyunsaturated fatty acid intake. Br J Nutr 1997; 78: 223-236.

14 Amiano P, Dorronsoro M, M de Renobales, Ruiz de Gordoa JC, Irigoien I, the EPIC Group of Spain. Very-long-chain ω-3 fatty acids as markers for habitual fish intake in a population consuming mainly lean fish: the EPIC cohort of Gipuzkoa. Eur J Clin Nutr 2001; 55: 827-832.

15 Oddy WH, Sherriff JL, Kendall GE et al. Patterns of fish consumption and level of serum phospholipid very-long-chain omega-3 fatty acids in children with and without asthma, living in Perth, Western Australia. Nutr Diet 2004; 61: 30-37.

16 Yep YL, Li D, Mann NJ, Bode O, Sinclair AJ. Bread enriched with microencapsulated tuna oil increases plasma docosahexaenoic acid and total omega-3 fatty acid in humans. Asia Pac J Clin Nutr 2002; 11: 285-291.

17 Gillingham LG, Caston L, Leeson S, Hourtovenko K, Holub BJ. The effects of consuming docosahexaenoic acid (DHA)-enriched eggs on serum lipids and fatty acid composition in statin-treated hypercholesterolemic male patients. Food Res International 2005; 38: 1117-1123.

18 Baró L, Fonallá J, Peña JL et al. n-3 fatty acids plus oleic acid supplemented milk reduces total and LDL cholesterol, homocysteine and levels of endothelial adhesion molecules in healthy humans. Clin Nutr 2003; 22: 175-182.

19 Garg ML, Blake RJ, Clayton E et al. Consumption of an n-3 polyunsaturated fatty acid-enriched dip modulates plasma lipid profile in subjects with diabetes type II. Eur J Clin Nutr 2007; 61: 1312-1317.

20 Murphy K, Meyer BJ, Mori TA et al. Impact of foods enriched with omega-3 long chain polyunsaturated fatty acids on erythrocyte omega-3 levels and cardiovascular risk factors. Br J Nutr 2007; 97: 749-757.

21 Australian Government, Department of Health and Ageing, National Health and Medical Research Council (2005) Food for health, dietary guidelines for Australian, a guide to healthy eating. Available at: http://www.nhmrc.gov.au/_files_nhmrc /publications/attachments/n31.pdf: accessed on February 2012.

22 Colquhoun D, Ferreira-Jardim A, Udell T, Eden B, the Nutrition and Metabolism Committee of the Heart Foundation. Review of evidence fish, fish oils, n-3 pollyunsaturated fatty acids and cardiovascular health. NHFA 2008. Available at: www.heartfoundation.org.au: accessed on December 2011.

23 National Health Foundation of Australia. Position statement, fish, fish oils, n-3 polyunsaturated fatty acids and cardiovascular health. NHFA 2008.

Chapter 3. Fish and seafood survey

24 Clayton EH, Hanstock TL, Watson JF. Estimated intakes of meat and fish by children and adolescents in Australia and comparison with recommendations. Br J Nutr 2009; 101:1731-1735.

25 Meyer BJ, Kolanu N. Australian children are not consuming enough long-chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 2011; 27: 1136- 1140.

26 Lovegrove JA, Brooks CN, Murphy MC, Gould BJ, Williams CM. Use of manufactured foods enriched with fish oils as a means of increasing long chain n-3 polyunsaturated fatty acid intake. Br J Nutr 1997; 78: 223-236.

27 Patch CS, Tapsell LC, Mori TA et al. The Use of Novel Foods Enriched with Long- Chain n-3 Fatty Acids to Increase Dietary Intake: A Comparison of Methodologies Assessing Nutrient Intake. J Am Diet Assoc 2005; 105: 1918-1926.

28 Birch LL. Preschool children’s food preferences and consumption patterns. J Nutr Educ. 1979;11:189-192.

29 Skinner JD, Carruth BR, Bounds W, Ziegler PJ. Children’s food preferences: A longitudinal analysis. J Am Diet Assoc 2002; 102: 1638-1647.

30 Kral TVE, Rauh EM. Eating behaviour of children in the context of their family environment. Phy and Behav 2010; 100: 567-573.

31 Savage JS, Fisher JO, Birch LL. Parental influence on eating behaviour: conception to adolescence. J Law Med Ethics 2007; 35: 22-34.

32 Imm P, Knobeloch L, Anderson HA. Maternal recall of children’s consumption of commercial and sport-caught fish: Findings from a multi-state study. Environ Res 2006; 103: 198-204.

33 Birch LL. Development of food preferences. Annu Rev Nutr 1999; 9: 41-62.

34 Gilman M, Rifas-Shiman S, Frazier L et al. Family dinner and diet quality among older children and adolescents. Archives Family Med 2000; 9: 235–240.

35 Jenkin S, Horner SD. Barriers that influence eating behaviours in adolescents. J Paediatr Nurs 2005; 20: 258-267.

36 Leek S, Muddock S, Foxall G. Situational determinants of fish consumption. Br J Food 2000; 102: 18-39.

37 Verbeke W, Vackier I. Individual determinants of fish consumption: application of the theory of planned behaviour. Appetite 2005; 44: 67-82.

38 Trondsen T, Scholderer J, Lund E, Eggen AE. Perceived barriers to consumption of fish among Norwegian women. Appetite 2003; 41: 301-314.

Chapter 3. Fish and seafood survey

39 Prell H, Berg C, Jonsson L. Why don’t adolescents eat fish? Factors influencing fish consumption in school. Scandinavian J Articles 2002; 46: 184-191.

40 Pliner P, Salvy SJ. Food neophobia in humans. In R Shepherd , M Raats (Eds). The psychology of food choice (pp. 75-92). Oxfordshire: CABI, 2006.

41 Knaapila A, Silventoinen K, Broms U et al. Food neophobia in young adults: genetic architecture and relation to personality, pleasantness and use frequency of foods, and body mass index – A twin study. Behav Genet 2011; 41: 512-521.

42 Fisheries Research and Development Corporation. A study of the retail sale and in home consumption of seafood in Sydney. Canberra, Commonwealth Dept of Agriculture, Fishery and Forestry, 1998.

43 Fisheries Research and Development Corporation. A study of seafood consumption in Perth. Canberra, Commonwealth Dept of Agriculture, Fishery and Forestry, 1999.

44 Fisheries Research and Development Corporation. Community perceptions of the sustainability of the fishing industry in Australia. April 2011.

45 Howe P, Buckley J, Meyer B. Red meat: a source of long chain omega-3. Nutr Diet 2007; 64: S135-139.

46 Patch CS, Tapsell LC, Williams PG. Overweight consumers’ salient beliefs on omega-3- enriched functional foods in Australia’s Illawarra region. J Nutr Educ Behav 2005; 37: 83-89.

47 Cox DN, Evans G, Lease HJ. Predictors of Australian consumers’ intentions to consume conventional and novel sources of long-chain omega-3 fatty acids. Public Health Nutr 2008; 11: 8-16.

48 Mozaffarian D, Rimm EB. Fish intake, contaminants, and human health evaluating the risks and the benefits. J Am Med Assoc 2006; 296: 1885-1899.

49 Food and Agriculture OrganizationWorld Health Organization. Report of the joint FAO/WHO expert consultation on the risk and benefits of fish consumption, 25-29 January 2010, Rome, Italy. FAO Fisheries and Aquaculture Report 2011, No 978.

50 Zhou SJ, Gibson RA, Gibson RS, Makrides M. Nutrient intakes and status of preschool children in Adelaide, South Australia. MJA 2012; 196: 696-700.

51 Burrows T, Berthon B, Garg ML, Collins CE. A comparative validation of a child food frequency questionnaire using red blood cell membrane fatty acids. Eur J Clin Nutr 2012; 66: 852-829.

52 Myrland O, Trondsen T, Johnston R, Lund E. Determinants of seafood consumption in Norway: lifestyle, revealed preferences and barriers to consumption. Food Quality and Pref 2000; 11:169-188.

Chapter 3. Fish and seafood survey

53 Wang Y, Beydoun MA, Li J, Liu Y, Moreno LA. Do children and their parents eat a similar diet? Resemblance in child and parental dietary intake: systemic review and meta-analysis. J Epid Comm Health 2011; 65: 177-189.

54 Scaglioni eS, Arrizza C, Vecchi F, Tedeschi S. Determinants of children’s eating behaviour. Am J Clin Nutr 2011; 94: 206S-11S.

55 National Health Foundation of Australia. Omega-3 levels in fish and seafood. NHFA 2008. http://www.heartfoundation.org.au/SiteCollectionDocuments/Omega3levelsinfishand seafood.pdf: access on 31 March 2012

56 Gladyshev MI, Sushchik NN, Gubanenko GA, Demirchieva SM, Kalachova GS. Effect of way of cooking on content of essential polyunsaturated fatty acids in muscle tissue of humpback salmon (Oncorhynchus gorbuscha). Food Chem 2006; 96: 446-451.

57 Candella M, Astiasarán I, Bello J. Deep-fat frying modifies high-fat fish lipid fraction. J Agri Food Chem 1998; 46: 2793-2796.

58 Candela M, Astiasarán I, Bello J. Effects of frying and warmholding on fatty acids and cholesterol of sole (Solea solea), codfish (Gadus morrhua) and hake (Merluccius merluccius). Food chem 1997; 58: 227-231.

59 Krutulyte R, Grunert KG, Scholderer J et al. Motivational factors for consuming omega-3 PUFAs: An exploratory study with Danish consumers. Appetite 2008; 51: 137-147.

60 Cox DN, Evans G, Lease HJ. Predictors of Australian consumers’ intention to consume conventional and novel sources of long-chain omega-3 fatty acids. Public Health Nutr 2007; 11: 8-16.

61 Brunsø. Motives, barriers and quality evaluation in fish consumption situations: Exploring and comparing heavy and light users in Spain and Belgium. Br Food J 2009; 7: 699-716.

62 Aslin HJ, Byron IG. Community perceptions of fishing: implication for industry image, marketing and sustainability. Fisheries Research and Development Corporation and Bureau of Rural Statistic 2003.

Chapter 4. Dietary modeling of n-3 LCPUFA

Chapter 4

The majority of this section is the substantive content of the work submitted to Nutrition.

The findings of this study were also presented in the 2nd International Food Studies at the University Illinois at Urbana Champaign, USA (oral presentation) and the 36th Nutrition Society of Australia, Wollongong 2012 (oral presentation). Abstract was published in the Proceedings of the Nutrition Society of Australia 36, 45.

Statement

As the primary supervisor, I, Barbara J Meyer, declare that the greater part of the work in this paper is attributed to the candidate, Setyaningrum Rahmawaty.

In the manuscript, Setyaningrum contributed to study design and was primarily responsible for data collection, data analysis, data interpretation and writing up the manuscript.

A/Prof Barbara J Meyer (Main Supervisor)

Chapter 4. Dietary modeling of n-3 LCPUFA

Effect of replacement of bread, egg, milk and yogurt with equivalent n-3 enriched foods on n-3 LCPUFA intake of Australian children

(Submitted to Nutrition)

Setyaningrum Rahmawatya,b, Philippa Lyons-Wallc, Marijka Batterhamd, Karen Charltonb, Barbara J Meyera,b

aMetabolic Research Centre, University of Wollongong, Wollongong, New South Wales, Australia bSchool of Health Sciences, University of Wollongong, Wollongong, New South Wales, Australia cSchool of Exercise and Health Sciences, Edith Cowan University, Joondalup, Western Australia, Australia. dStatistical Consulting Service, University of Wollongong, Wollongong, Wollongong, New South Wales, Australia

Corresponding author. Tel.: +61 (0)2 4221 3459, Fax: +61 (0)2 4221 5945 Email address: [email protected] (B.J. Meyer)

SR performed the statistical analysis and interpretation of the data and drafted the manuscript. PLW participated in study design and assisted with the first draft of the manuscript, KC co-authored the manuscript and was involved in conceptualization of the study, MB provided statistical consultation and BJ was the originator of the idea of the study, participated in study design and assisted with the draft of the manuscript. All authors contributed to the manuscript, read and approved the final manuscript.

Abstract

In countries with traditionally low fish consumption such as Australia, foods enriched with omega-3 long chain polyunsaturated fatty acids (n-3 LCPUFA) may play a role in meeting n-3 LCPUFA intakes for optimal health. The aim of the present study was to assess the effect of replacement of bread, egg, milk and yogurt with n-3 LCPUFA enrichment of these foods on total n-3 LCPUFA intake in Australian children’s diets. Dietary modelling was undertaken using survey data from a nationally representative sample of 4487 children (2249 boys, 2238 girls) aged 2-16 years in whom the Multiple Source Method (MSM) was used to estimate usual n-3 LPUFA intakes distributions from 2 x 24-h dietary recalls, corrected for within-person variation and fifteen models were constructed. The adjusted mean ± SD and median inter quartile range (IQR) usual dietary intakes of n-3 LCPUFA gradually increased from 2.5 ± 0.8 to 7.1 ± 4.9 mg/d and 2.3 (1.9-2.9) to 5.4 (3.6-9.2), respectively, after the modelling (p = 0.001 for each model). Median (IQR) intake of total n-3 LCPUFA’s in non-fish eaters and fish eaters

100

Chapter 4. Dietary modeling of n-3 LCPUFA

was 1.4 (0.8-2.3) and 2.3 (1.0-6.1) mg/d, respectively which increased by 3 fold to 4.3 (2.6-7.8) and 7.5 (3.9-13) mg/d, respectively after replacement of all four items. Consumption of n-3 enriched foods resulted in improved n-3 LCPUFA intakes in Australian children, without major changes to their current food habits. However, fish consumption is still the most effective strategy to increase n-3 LCPUFA intake.

Keywords: dietary modelling, n-3 enriched food, n-3 LCPUFA intakes, children, Australia

Introduction

Cardiovascular disease (CVD) is the leading cause of death in Australia(1). Ninety percent of Australian adults have at least one modifiable CVD risk factor and 25% have three or more modifiable risk factors(2). The progression of CVD can be delayed by lifestyle choices(3), and early intake of n-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) has been shown to have beneficial effects on the prevention of CVD in later life(4,5).

Improved clinical outcomes for cardiovascular risk have been demonstrated in randomized double-blind controlled studies in children aged between 2-18 years following n-3 LCPUFA supplementation. These include reductions in biomarkers of CVD risk in healthy children(6), modulation of vascular function and inflammatory markers in obese adolescents(7), and improvement in vascular health in hyperlipidemic children at risk for early heart disease(8).

De novo synthesis of n-3 LCPUFA, including eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA) and docosahexaenoic acid (DHA) is inefficient in (9) humans, especially for DHA , and these n-3LCPUFAs therefore need to be obtained from dietary sources, mainly from fish with smaller amounts supplied by meat and egg products(10). Moderate amounts of n-3 LCPUFA are also found in commercially available foods that have been enriched with n-3 LCPUFA, including selected brands of bread, cereal, milk, yogurt and eggs(11). For cardiovascular health, the National Heart Foundation of Australia (NHFA) recommends that children should follow the adult guideline of about 500 mg/d for EPA plus DHA, or 2 or 3 serves per week of 150 g fish, preferably oily fish, combined with food and drinks enriched with marine n-3 fatty

101

Chapter 4. Dietary modeling of n-3 LCPUFA

acids(12,13). In Australia, the National Health and Medical Research Council (NHMRC) has established the nutrient reference value for n-3 LCPUFA in children less than 14 years based on the adequate intake (AI) or observed median intake from the National Dietary Survey(14). For children aged 14-16 years, a suggested dietary target (SDT) recommended for prevention of chronic diseases, has been set at 610 and 430 mg/d for boys and girls, respectively based on the observed 90th percentile of the population (14) intake of Australian children in the 1995 National Dietary Survey data . Meyer and Kolanu (2011) have extrapolated SDT for children younger than 14 years by adjusting for energy intakes, by sex and age group(15).

The number of foods with added n-3 LCPUFA is increasing, particularly in western countries such as Australia(11) where milk, cereal and bread enriched with n-3 (16) LCPUFA are potential contributors to n-3 LCPUFA intakes . However less than 7 % of Australian children are reported to consume foods enriched with n-3 LCPUFA(15), which has been proposed as a strategy to increase n-3 LCPUFA intake at a population level(16,17). The effectiveness of substituting foods with n-3 LCPUFA enriched alternatives has not yet been investigated in the population of Australian children. Previous prospective studies have demonstrated that consumption of foods enriched with n-3 LCPUFA improves n-3 LCPUFA intake (17,18) and CVD risks(18) but that this strategy requires individual behavior changes which may not be maintained over the long term. The current study aimed to assess whether replacement of bread, milk, egg and yogurt with n-3 LCPUFA enriched brands would enable Australian children to meet the recommendation for n-3 LCPUFA for optimal health, without the need to change individual food behaviours.

Material and methods

Subjects

Dietary data were available from children participating in the Australian National Children’s Nutrition and Physical Activity Survey (Children’s Survey) which was conducted between 22 February and 30 August 2007. The survey was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the National Health and Medical Research Council (NHMRC) of Australia registered ethics committees of the Commonwealth

102

Chapter 4. Dietary modeling of n-3 LCPUFA

Scientific and Industrial Research Organization (CSIRO) and University of South Australia(19). The data used in this study were obtained with permission from the Australian Social Science Data Archive(20). The protocol of the survey has been reported elsewhere(21). Participants in the survey were children aged between 2 and 16 years (n = 4487) randomly selected using random digit dialing from all Australian states and territories in metropolitan, rural and remote areas.

Dietary data

Two 24-hour dietary recalls were available for each subject, collected using a standardized 24-hour dietary recall methodology during a computer assisted personal interview (CAPI) and a computer assisted telephone interview (CATI)(21). Analyses of the 24-hour dietary recall data into daily nutrient intake were conducted using the Australian nutrient composition database (AUSNUT) 2007 developed specifically for the 2007 Children’s Survey(19). The intake of total n-3 LCPUFA was determined by summing individual intakes of EPA, DPA and DHA. The usual daily dietary n-3 LCPUFA intake, corrected for within-person variation, was estimated using data from the two 24-hour dietary recalls and applying the Multiple Source Method (MSM)(22). With this method, the total variance was adjusted for the intra-individual variances due to day-to-day variability(22). Special handling information appeared in the result section during the analysis by the MSM, where only positive and negative values for skewness of variable were encountered during Box-Cox transformation. “When only positive values for skewness are countered during Box-Cox transformation of residual, MSM will use the best parameter estimates that lead to the residual distribution skewness closest to zero. When only negative values for skewness are encountered during Box- Cox transformation of residuals, MSM will use the parameters that lead to the best skewness value for the residual distribution (which is defined as being is closest to zero) to determining Lambda, W and skewness for the Box-Cox transformation” 22.

Dietary modelling

The dietary modelling was conducted by incrementally replacing all types of bread, egg, milk and yogurt items in the two 24-hour food recalls for each child with the corresponding n-3 LCPUFA enriched products. The dietary modelling was run in four stages: (1) replacement with one type n-3 LCPUFA enriched food (bread, eggs, milk or

103

Chapter 4. Dietary modeling of n-3 LCPUFA

yogurt); (2) replacement with two types n-3 LCPUFA enriched foods (bread + egg, bread + milk, bread + yogurt or bread + egg); (3) replacement with three types n-3 LCPUFA enriched foods (bread + milk + yogurt, bread + yogurt + egg, milk + yogurt + egg or bread + egg + milk) and (4) replacement with four types n-3 LCPUFA enriched foods (bread + milk + yogurt + egg). This procedure generated a total of 15 models of dietary n-3 LCPUFA intake. The content of EPA and DHA in enriched items was obtained from a survey of nutrient content on food labels conducted at three major supermarket chains (Woolworths, Coles and Aldi) in Wollongong and Sydney in December 2009 (Table 1A). DPA was not included, as this was not used by manufactures to enrich food items. The modeled average intakes of total n-3 LCPUFA were compared to the AI and the SDT, where the NHMRC of Australia has defined AI as “The average daily nutrient intake level based on observed or experimentally- determined approximations or estimates of nutrient intake by a group (or groups) of apparently healthy people that are assumed to be adequate” and the SDT as “A daily average intake from food and beverages for certain nutrients that may help in prevention of chronic disease”(14). Hence, the AI reflects the median intakes of the population and is not a recommended intake per se, while SDTs are target intakes for optimal health, rather than prevention of deficiency states.

The cost of the meal was calculated with the assumption that the amount (g) of bread, egg, milks and yogurt consumed by each child was enriched with n-3 LCPUFA. The differences in price between bread, egg, milk and yogurt with and without n-3 LCPUFA enrichment were used in calculating the cost of meals. The price of these foods was obtained from the supermarkets survey as described above.

The Microsoft Excel VLOOKUP function was used to run the replacement of the n-3 LCPUFA enriched foods in each model as well as the cost of meals. Briefly, the master database of foods in the AUSNUT 2007 specifically used to calculate EPA, DPA, DHA and n-3 LCPUFA and the data of foods consumed by each child from two 24-hour dietary recalls were provided in an Excel spread. The nutrient content of EPA, DPA and DHA in bread, egg, milk and yogurt items in the Children’s Survey was replaced with the corresponding n-3 LCPUFA enriched items in each model. The food identity (Food ID) in the AUSNUT 2007 was used as a key to merge the EPA, DHA

104

Chapter 4. Dietary modeling of n-3 LCPUFA

and n-3 LCPUFA value to the data consumed by the total group of children in each model.

Statistical analysis

The statistical analysis was carried out using Statistical Package for the Social Sciences (SPSS) software version 17.0, Chicago IL, USA. Intakes of EPA, DHA and total n-3 LCPUFA (mg/d) are presented as mean ± SD and median (IQR). Intakes of n-3 LCPUFA were determined by fish eater status, based on consumption of at least one serve of fish (yes/no) from the two-24 hour dietary recalls. The data were tested for normality using the Kolmogorov-Smirnov test and as the data were skewed, the differences in n-3 LCPUFA intake before and after replacement were assessed using non-parametric tests. The Friedman test was used to compare the difference between the mean ranks of the related models. Post-hoc analysis with Wilcoxon Signed-Rank tests were conducted to locate significant differences; a significance level using the Bonferroni adjustment for multiple comparison correction was set at p<0.003.

Results

Dietary n-3 LCPUFA intake and cost of meals before and after modelling

In our study, the average amount of bread, egg, milk and yogurt consumed by the children varied between age groups and ranged from 52 to 84 g/d for bread, 22 to 35 g/d for egg, 114 to173 mL/d for milk, and 56 to 86 g/d for yogurt (Table 1B). These portion sizes resulted in a range of intakes for n-3 LCPUFA: bread: 22.9 to 37.0 mg/d; egg: 44.0 to 70 mg/d; milk: 13.7 to 20.8 mg/d; and yogurt: 37.5 to 57.6 for children aged 2-3, 4-8, 9-13 and 14-16 years, respectively. The mean cost of meals in children who consumed bread, egg, milk and yogurt increased by AUD 45 cents (€ 0.36 or US$ 0.47) with replacement of all of the potential corresponding foods (model 15).

105

Chapter 4. Dietary modeling of n-3 LCPUFA

Table 1A. Omega-3 LCPUFA content in n-3 LCPUFA enriched bread, egg, milk and yogurt

EPA DPA DHA Total n-3 LCPUFA* (mg/100g) (mg/100g) (mg/100g) (mg/100g) Bread (n = 4) 8a 0 36 44 Egg (n = 3) 60 0 140 200 Milk (n = 1) 0.3 0 11.7 12.0 Yogurt (n = 4) 0 0 67 67

From nutrient content listed on item labels, obtained from a supermarket survey conducted by N Kolanu and S Rahmawaty in Wollongong and Sydney, December 2009 (unpublished data) aAverage amount, *Sum EPA+DPA+DHA, the author’s calculation

Table 1B. Intake of bread, egg, milk and yogurt in consumers

Mean ± SD intake (in consumers) 2-3 y 4-8 y 9-13 y 14-16 y All ages (n, %) (n, %) (n, %) (n, %) Bread 52 ± 42a 58 ± 47 72 ± 63 84 ± 74 63 ± 59 (g/d) (1035, 24)b (1188, 28) (1052, 25) (1018, 24) (4293) Egg 22 ± 19 28 ± 21 31 ± 26 35.2 ± 29 7 ± 17 (g/d) (254, 25) (271, 27) (242, 24) (250, 25) (1017) Milk 114 ± 67 139 ± 89 156.3 ± 103 173 ± 130 137 ± 104 (g/d) (1051, 25) (1169, 28) (1041, 24) (982, 23) (4243) Yogurt 56 ± 31 57 ± 29 72 ± 44 86 ± 60 21 ± 37 (g/d) (541, 37) (459, 31) (255, 17) (210, 14) (1465) aMean ± SD intake (in consumer during the two days of survey) bNumber of consumers, percentage of consumers within the all ages

Median (IQR) of total n-3 LCPUFAs for the whole group adjusted for the intra- individual variances was 2.3 (1.9-2.9) mg/d and this increased by 3.1 mg (115 %) to 5.4 (3.6-9.2) mg/d after replacement of all four items (model 15). Corresponding intakes of individual fatty acids also increased. With EPA, the greatest increase occurred with replacement of milk (model 3) and further small increases occurred with replacement of bread, yogurt and egg (model 5, 12). With DHA, the greatest increases occurred with replacement of yogurt (model 4) and milk (model 3) or combination of both (model 8) and further with additional replacement of bread and egg (model 11 and 15). Statistically significant differences (p = 0.001) were observed between intakes of EPA, DPA, DHA and total n-3 LCPUFA between actual diet (before replacement) and each level of replacement, except for EPA intake in model 4 (Table 2).

106

Chapter 4. Dietary modeling of n-3 LCPUFA

Table 2. Modelling of usual daily intake of EPA, DPA, DHA and total n-3 LCPUFA before and after replacement with n-3 enriched foods for all children (n = 4487)a EPA DHA Total n-3 LCPUFAb n-3 enriched foods Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Baseline (actual diet) 0.9 ± 1.9 0.4 (0.2-0.9) 1.0 ± 2.5 0.4 (0.3-0.7) 2.5 ± 0.8 2.3 (1.9-2.9) 1 item Model 1 (bread) 1.0 ± 2.5 0.5* (0.2-1.1) 1.3 ± 2.4 0.7* (0.5-1.1) 2.9 ± 0.8 2.7* (2.3-3.2) Model 2 (egg) 1.0 ± 2.1 0.5* (0.2-1.1) 1.1 ± 3.1 0.5* (0.3-0.8) 2.6 ± 1.1 2.4* (1.9-3.1) Model 3 (milk) 0.8 ± 0.8 0.6* (0.3-1.0) 3.3 ± 4.4 1.5* (1.1-2.7) 5.3 ± 3.9 4.0* (2.6-6.7) Model 4 (yogurt) 0.9 ± 1.9 0.4 (0.2-1.0) 3.1 ± 4.3 1.8* (1.5-2.2) 4.2 ± 8.2 1.9* (0.8-4.5) 2 items

Model 5 (bread, egg) 1.0 ± 0.4 0.9* (0.8-1.0) 1.5 ± 2.8 0.7* (0.5-1.2) 3.0 ± 1.1 2.8* (2.3-3.6) Model 6 (bread, milk) 0.9 ± 0.9 0.6* (0.4-1.0) 3.6 ± 3.9 1.8* (1.4-3.1) 5.6 ± 3.9 4.3* (2.9-7.3) Model 7 (bread, yogurt) 1.0 ± 2.5 0.5* (0.2-1.1) 2.9 ± 4.2 1.6* (1.1-2.6) 4.7 ± 1.3 4.5* (3.7-5.4) Model 8 (milk, yogurt) 0.8 ± 0.8 0.6* (0.3-1.0) 4.6 ± 5.1 2.4* (1.7-4.6) 6.6 ± 4.6 5.0* (3.4-8.6) Model 9 (milk, egg) 0.9 ± 0.9 0.6* (0.3-1.0) 3.4 ± 4.6 1.6* (1.1-2.9) 5.5 ± 4.2 4.1* (2.7-7.0) Model 10 (yogurt, egg) 1.0 ± 2.1 0.5* (0.2-1.1) 3.3 ± 4.8 1.9* (1.5-2.6) 4.4 ± 0.3 4.4* (4.2-4.6) 3 items Model 11 (bread, milk, yogurt) 1.0 ± 0.6 0.8* (0.6-1.2) 4.8 ± 4.8 2.7* (1.9-5.3) 6.9 ± 4.7 5.3* (3.6-9.0) Model 12 (bread, yogurt, egg) 1.0 ± 0.4 0.9* (0.8-1.0) 3.1 ± 4.6 1.7* (1.1-2.8) 4.9 ± 1.7 4.5* (3.6-5.8) Model 13 (milk, yogurt, egg) 0.9 ± 0.9 0.6* (0.3-1.0) 4.7 ± 5.4 2.5* (1.7-4.7) 6.8 ± 4.9 5.2* (3.4-5.2) Model 14 (bread, egg, milk) 0.9 ± 1.0 0.7* (0.4-1.1) 3.7 ± 4.1 1.9* (1.4-3.4) 5.8 ± 4.1 4.4* (2.9-7.5) 4 items Model 15 (bread, egg, milk, yogurt)c 1.0 ± 1.0 0.7* (0.4-1.1) 4.9 ± 4.9 2.8* (2.0-5.4) 7.1 ± 4.9 5.4* (3.6-9.2) aEstimated from two 24-hour dietary recalls using the Multiple Source Method (MSM) software to account for day today variation(22) EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; total n-3 LCPUFA, sum EPA+DPA+DHA; SD, standard deviation; IQR, Interquartile range; cThe average cost of meal in children who ate bread, egg, milk and yogurt increased AUD 45 cent per day; *Significant difference between the baseline (actual intake) and intake from each model (p = 0.001) using Post-hoc analysis with Wilcoxon Signed-Rank Tests with a Bonferroni correction applied

107

Chapter 4. Dietary modeling of n-3 LCPUFA

Twenty percent of all children consumed fish on at least one of the days of the survey. Median (IQR) baseline intake of total LCPUFA’s in non-fish eaters was 1.4 (0.8-2.3) mg/d and this increased by 2.9 mg (207 %) to 4.3 (2.6-7.8) mg/d after replacement of all four items (Table 3A). Median (IQR) baseline intake of total n-3 LCPUFA’s in fish eaters was 2.3 (1.0-6.1) mg/d and this increased by 5.2 mg (226 %) to 7.5 (3.9-13) mg/d after replacement of all four items (Table 3B).

DHA intake in non-fish eating children increased 23-fold, as compared to the 7- fold increase in the fish eater group, due to 6-fold higher starting levels in the latter. In non-fish eaters, DHA increased from 0.1 to 2.3mg/d, whilst in the fish eaters group increased from 0.6 to 4.4mg/d. This suggests that children who reported consuming fish were also consuming higher quantities of bread, milk, eggs and yoghurt, therefore these enriched foods contributed more DHA (i.e. 3.8 mg increase) than in non-fish eating children (2.2 mg).

Distribution of n-3 LCPUFA intake and comparison between modelled diet and nutrient reference values for n-3 LCPUFA

As a result of the modelling, the distribution of intakes adjusted for intra-individual variances shifted towards higher intakes of n-3 LCPUFA (Figure 1). Substitution of n-3 enriched milk for un-enriched milk (model 3) virtually doubled the n-3 LCPUFA intakes and shifted the curve to the right (figure 1). The addition of the other n-3 enriched foods (bread, eggs and yoghurt) to the enriched milk (model 15) shifted the curve further to the right (figure 1).

Based on average intake of n-3 LCPUFA (non-adjusted intra-individual variances), there was a greater proportion of children reaching the AI for total n-3 LCPUFA intake (28.1 % to 43.3 %) and minimal change in the proportion of children who reached the SDT (3.8 % to 4.2 %) (Online Supplementary Figure 1). The percentage of children that reach the AI for n-3 LCPUFA before and after modeling increased from 13 % to 31 % in the non- fish and 85 % to 91 % in the fish eater groups, while for SDT a minimal increase from 0.05 % to 0.1 % in the non-fish eater and 18 % to 20 % in the fish eater groups were demonstrated.

108

Chapter 4. Dietary modeling of n-3 LCPUFA

Table 3A. Modelling of usual daily intake of EPA, DPA, DHA and total n-3 LCPUFA before and after replacement with n-3 enriched foodsa (n = 3554)b in non-fish eaterc Non-fish eater (n = 3554) EPA DHA Total n-3 LCPUFA Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Baseline (actual diet) 0.7 ± 0.5 0.5 (0.4-0.9) 0.4 ± 1.6 0.1 (0.1-0.3) 1.9 ± 1.8 1.4 (0.8-2.3) 1 item Model 1 (bread) 0.7 ± 0.6 0.5 (0.4-0.8) 0.7 ± 1.5 0.3* (0.2-0.6) 2.2 ± 1.5 1.8* (1.2-2.8) Model 2 (egg) 0.8 ± 0.6 0.6* (0.4-1.0) 0.5 ± 2.4 0.1 (0.1-0.4) 2.0 ± 2.1 1.4 (0.8-2.4) Model 3 (milk) 0.7 ± 0.9 0.4 (0.2-0.7) 2.7 ± 3.6 1.3* (0.8-2.3) 4.4 ± 4.0 3.1* (1.9-5.5) Model 4 (yogurt) 0.7 ± 0.5 0.5 (0.4-0.9) 1.9 ± 3.7 0.9* (0.5-1.9) 3.4 ± 3.0 2.5* (1.5-4.3) 2 items Model 5 (bread, egg) 0.8 ± 0.7 0.6* (0.4-0.9) 0.8 ± 2.0 0.3* (0.2-0.7) 2.4 ± 1.9 1.9* (1.2-3.0) Model 6 (bread, milk) 0.7 ± 0.9 0.5 (0.2-0.8) 3.0 ± 3.5 1.5* (1.0-2.7) 4.7 ± 3.9 3.4* (2.1-6.0) Model 7 (bread, yogurt) 0.7 ± 0.6 0.5 (0.4-0.8) 1.8 ± 3.4 0.9* (0.4-1.8) 3.6 ± 3.1 2.8* (1.7-4.5) Model 8 (milk, yogurt) 0.7 ± 0.9 0.4 (0.2-0.7) 3.9 ± 4.6 2.0* (1.3-3.7) 5.5 ± 4.8 4.0* (2.4-7.0) Model 9 (milk, egg) 0.7 ± 1.1 0.4 (0.2-0.8) 2.8 ± 4.0 1.3* (0.8-2.4) 4.6 ± 4.3 3.2* (1.9-5.8) Model 10 (yogurt, egg) 0.8 ± 0.6 0.6* (0.4-1.0) 2.1 ± 4.2 0.9* (0.5-2.1) 3.6 ± 3.5 2.5* (1.5-4.5) 3 items Model 11 (bread, milk, 0.7 ± 0.9 0.5 (0.2-0.8) 4.1 ± 4.6 2.3* (1.5-4.2) 5.8 ± 4.9 4.2* (2.6-7.6) yogurt) Model 12 (bread, yogurt, egg) 0.8 ± 0.7 0.6* (0.4-0.9) 2.0 ± 3.8 0.9* (0.4-1.9) 3.9 ± 3.5 2.8* (1.7-4.8) Model 13 (milk, yogurt, egg) 0.7 ± 1.1 0.4 (0.2-0.8) 4.0 ± 4.9 2.1* (1.3-3.9) 5.7 ± 5.2 4.1* (2.4-7.4) Model 14 (bread, egg, milk) 0.8 ± 1.1 0.5 (0.2-0.9) 3.1 ± 3.8 1.6* (1.0-2.8) 4.8 ± 4.3 3.5* (2.1-6.2) 4 items Model 15 (bread, egg, milk, 0.8 ± 1.1 0.5 (0.2-0.9) 4.2 ± 4.8 2.3* (1.5-4.4) 6.0 ± 5.3 4.3* (2.6-7.8) yogurt) aModelled diet by replacing bread, egg, milk and yogurt with n-3 enriched for these foods in the actual diet of each child for all children (as note in Table 1) bEstimated from two 24-hour dietary recalls using the Multiple Source Method (MSM) software to account for day today variation(22) cFish eater, ate fish from the two 24-hour recalls; Non-fish eater, did not eat fish from the two 24-hour recalls; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; total n-3 LCPUFA, sum EPA+DPA+DHA; SD, standard deviation; IQR, Interquartile range; *Significant difference between the baseline (actual intake) and intake from each model (p = 0.001) using Post-hoc analysis with Wilcoxon Signed-Rank Tests with a Bonferroni correction applied

109

Chapter 4. Dietary modeling of n-3 LCPUFA

Table 3B. Modelling of usual daily intake of EPA, DPA, DHA and total n-3 LCPUFA before and after replacement with n-3 enriched foodsa (n = 933)b in fish eaterc Fish eater (n = 933) EPA DHA Total n-3 LCPUFA Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Baseline (actual diet) 2.7 ± 9.9 0.7 (0.2-2.1) 7.4 ± 40.0 0.6 (0.2-2.8) 7.1 ± 20.0 2.3 (1.0-6.1) 1 item Model 1 (bread) 2.6 ± 7.7 0.8* (0.3-2.3) 6.0 ± 28.0 0.9* (0.3-3.4) 7.6 ± 20.0 2.5* (1.1-6.8) Model 2 (egg) 2.7 ± 9.9 0.7 (0.2-2.1) 7.4 ± 40.0 0.6 (0.2-2.8) 7.1 ± 20.0 2.3 (1.0-6.1) Model 3 (milk) 1.8 ± 3.7 0.8* (0.4-1.7) 5.9 ± 12.0 2.5* (1.2-5.6) 8.9 ± 12.0 5.9* (2.8-10.7) Model 4 (yogurt) 2.7 ± 9.9 0.7 (0.2-2.1) 8.2 ± 31.0 1.4* (0.4-5.2) 10.0 ± 28.0 3.2* (1.3-9.3) 2 items Model 5 (bread, egg) 2.6 ± 7.7 0.8* (0.3-2.3) 6.0 ± 28.0 0.9* (0.3-3.4) 7.6 ± 20.0 2.5* (1.1-6.8) Model 6 (bread, milk) 1.8 ± 3.5 0.9* (0.4-1.9) 6.0 ± 10.0 3.0* (1.5-6.5) 9.1 ± 12.0 6.2* (3.2-11.4) Model 7 (bread, yogurt) 2.6 ± 7.7 0.8* (0.3-2.3) 7.2 ± 24.0 1.9* (0.6-5.7) 10.0 ± 27.0 3.5* (1.5-9.7) Model 8 (milk, yogurt) 1.8 ± 3.7 0.8* (0.4-1.7) 7.3 ± 12.0 3.7* (1.8-8.2) 10.0 ± 13.0 6.9* (3.5-12.8) Model 9 (milk, egg) 1.8 ± 3.7 0.8* (0.4-1.7) 5.9 ± 12.0 2.5* (1.2-5.6) 8.9 ± 12.0 5.9* (2.8-10.7) Model 10 (yogurt, egg) 2.7 ± 9.9 0.7 (0.2-2.1) 8.2 ± 31.0 1.4* (0.4-5.2) 10.0 ± 28.0 3.2* (1.3-9.3) 3 items Model 11 (bread, milk, yogurt) 1.8 ± 3.5 0.9* (0.4-1.9) 7.4 ± 11.0 4.4* (2.2-8.7) 10.0 ± 12.0 7.5* (3.9-13) Model 12 (bread, yogurt, egg) 2.6 ± 7.7 0.8* (0.3-2.3) 7.2 ± 24.0 1.9* (0.6-5.7) 10.0 ± 27.0 3.5* (1.5-9.7) Model 13 (milk, yogurt, egg) 1.8 ± 3.7 0.8* (0.4-1.7) 7.3 ± 12.0 3.7* (1.8-8.2) 10.0 ± 13.0 6.9* (3.5-12.8) Model 14 (bread, egg, milk) 1.8 ± 3.5 0.9* (0.4-1.9) 6.0 ± 10.0 3.0* (1.5-6.5) 9.1 ± 12.0 6.2* (3.2-11.4) 4 items Model 15 (bread, egg, milk, yogurt) 1.8 ± 3.5 0.9* (0.4-1.9) 7.4 ± 11.0 4.4* (2.2-8.7) 10.0 ± 12.0 7.5* (3.9-13) aModelled diet by replacing bread, egg, milk and yogurt with n-3 enriched for these foods in the actual diet of each child for all children (as note in Table 1) bEstimated from two 24-hour dietary recalls using the Multiple Source Method (MSM) software to account for day today variation(22) cFish eater, ate fish from the two 24-hour recalls; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; total n- 3 LCPUFA, sum EPA+DPA+DHA; SD, standard deviation; IQR, Interquartile range; *Significant difference between the baseline (actual intake) and intake from each model (p = 0.001) using Post-hoc analysis with Wilcoxon Signed-Rank Tests with a Bonferroni correction applied

110

Chapter 4. Dietary modeling of n-3 LCPUFA

4000

3500

3000

M0 (actual diet)

2500 M3 (replacing with milk enriched with n-3)

2000 M15 (replacing with bread + egg + milk + yogurt enriched with n-3)

1500 Number Number childrenof

1000

500

0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60

Usual daily dietary intake of total n-3 LCPUFA for all children (mg/d)a

Fig 1. The changes distribution of usual daily dietary intake of total n-3 LCPUFA of the Australian children’s diet (n=4487) before and after replacing bread, milk, yogurt and egg with enriched n-3 for these foods. aEstimated from two 24-hour dietary recalls using the Multiple Source Method (MSM) software to account for day today variation(22)

111

Chapter 4. Dietary modeling of n-3 LCPUFA

Overall, the percentage of total children that reach the AI for n-3 LCPUFA increased with the incremental inclusion of the four enriched items, while the percentage of children that met the SDT showed minimal change (Online Supplement Figure 2).

Discussion

This study is the first to assess the effectiveness of a dietary strategy designed to increase n-3 LCPUFA intake in Australian children by consuming enriched foods. The n-3 LCPUFA enriched foods available in supermarkets and the actual diet of each child were used as a starting point to develop the dietary modelling. This dietary modelling method could be used to appraise the effectiveness of a dietary recommendation to increase n-3 LCPUFA intake in the community. Food fortification can lead to relatively rapid changes in the specific nutritional status of a community, and is a cost-effective public health intervention, however the items need to be consumed in adequate amounts by a large proportion of target individuals in the population and the levels of fortification must be high enough to substantially increase the intakes(23). In this modelling scenario, all types of bread, egg, milk and yogurt consumed by the children were enriched with n-3 LCPUFA as these foods are currently commercially available in supermarkets in Australia. The children’s diets were modeled using the children’s actual habitual intakes regarding the serving size and frequency of intake of these foods, hence the children would not need to change their dietary behavior; however, they (or their parents) need to purchase these enriched foods which would cost an extra AUD 0.45 dollars per day.

The amount of bread, egg, milk and yogurt consumed by children in this study varied from individual to individual. These variations influenced the changes of their total n-3 LCPUFA intake as a result of the dietary modelling. Our data showed a gradual increase in the mean and median intakes of total n-3 LCPUFA (sum EPA+DPA+DHA) adjusted for intra-individual variances of 2.84 and 2.35 fold increase, respectively, by replacing bread, egg, milk and yogurt with n-3 LCPUFA enriched choices for these foods in the actual diet of each child. This improvement did not result in children reaching the Australian dietary targets for n-3 LCPUFA intake (SDT). Notably however was the magnitude of increase in median DHA in the non-fish eater consumers which was 3-fold higher compared to the increase observed in the fish 112

Chapter 4. Dietary modeling of n-3 LCPUFA

consumers. This suggests that consumption of n-3 LCPUFA enriched foods benefits non-fish eaters more so than fish eaters, however, the improvement after modelling was still below the SDT. This means that eating fish is still the most effective option to increase n-3 LCPUFA intake.

One limitation to recommending an increase in foods enriched with n-3 LCPUFA is the need for the food items to be consumed in large enough quantities to meet the recommendation of 500 mg/d(24). According to manufacturers’ data for the marine fish oil LCPUFA content of bread and milk, it requires consumption of 10 slices bread or 1 L of milk to achieve the recommendation for n-3 LCPUFA, which is unlikely to be widely implemented on a daily basis because of individual food preferences and socioeconomic or cultural factors. Moreover, it is contrary to the recommendation to eat a variety of nutritious foods(25). It has been reported that regular consumption of a variety of n-3 LCPUFA enriched foods, consumed over eight servings per day, providing between 50 and 150 mg EPA plus DHA per serving, increased the daily n-3 LCPUFA intake of Australian adults from 200 mg/d to 960 mg/d(18), nearly two-fold higher than the recommendation for n-3 LCPUFA of 500 mg/d(13). Consumption of 600 mL of milk per day providing 120 and 60 mg DHA and EPA, respectively, increased the daily intake of n-3 LCPUFA (EPA + DHA) of healthy children by 180 mg/d(6).

The present study was a conceptual model and did not assess biochemical markers of n-3 LCPUFA. However, other studies have reported that increases in plasma, platelet and mononuclear cell phospholipid content of n-3 LCPUFA can be achieved by consumption n-3 LCPUFA enriched foods without the simultaneous ingestion of supplements or a change in dietary habits(26). A review of nine controlled intervention studies concluded that consumption of n-3 LCPUFA enriched milk reduced total cholesterol, LDL-cholesterol and triglycerides in healthy people, individuals with increased risk factor, as well as patients with CVD(27). Therefore, consumption of n-3 LCPUFA enriched foods could provide a long term strategy to improve the chronically low intakes of n-3 LCPUFA in the community(28), particularly in countries with traditionally low fish consumption such as Australia(29) where median n-3 LCPUFA intake in the current study was only 29 mg/d and in a previous study of adults, only 121 (30) mg/d . It has been reported that daily consumption of low dose n-3 LCPUFA from microencapsulated tuna oil enriched bread providing 60 mg of n-3 LCPUFA per day (10

113

Chapter 4. Dietary modeling of n-3 LCPUFA

mg n-3 LCPUFA per slice of bread) significantly increased concentration of n-3 LCPUFA in the total plasma lipids (18 % rise) and phospholipids fraction (12 % rise) of Australian adults after three weeks intervention(31). Regular consumption of a variety of n-3 LCPUFA enriched foods providing between 50 and 150 mg EPA plus DHA per serving increased in EPA, DHA and n-3 LCPUFA concentration in erythrocytes by 82, 111 and 35 % and 53, 76 and 53 % at 3 and 6 months respectively, after supplementation(18). Additionally, this improvement was associated with reduction in CVD risks, including a positive association with arterial compliance and a negative association with serum C reactive protein and urinary 11-dehydro-Thromboxane B2 excretion(18). The concentration of EPA and DHA expressed as the percentage of total fatty acid (omega-3 index) increased from 4 to 7 % over 6 months study(18), placing these subjects in lower risk for cardiac death(32).

In our study, we assumed that the replacement with n-3 LCPUFA enriched foods would not negatively impact on other nutrients, particularly lipid profiles and body weight. Previous studies have shown an increase in n-3 LCPUFA enriched foods consumption had beneficial effects to lipid profiles in healthy volunteers(33-35) as well as in hypercholesterolemic subjects(36). Consumption of 600 mL/d of milk enriched with fish oils, minerals and vitamins resulted in no significant changes in body mass index (BMI) and waist circumference of healthy children(6). Similar findings have also been found with consumption of bread(31) as well as sausages and French onion dip enriched with n-3 LCPUFA(26). A number of intervention studies that have focused on change in dietary behaviour in order to achieve a daily intake of 1 to 1.8 g/d n-3 LCPUFA have resulted in weight gain in the participants(17,18). Since our study used actual food consumption data from a nationally representative sample of children to model scenarios of replacing foods with similar energy content, no impact on weight gain would be expected.

A concern of food enriched with n-3 is bioavailability and possible physiological effects in the systemic circulation or the target organ. It has reported that the food matrix or carrier food, and lipid structure can affect the bioavailability of n-3 LCPUFA(37). Various studies have investigated physiological responses to different types of foods that have been enriched with n-3 LCPUFA. Consumption of bread enriched with microencapsulated tuna oil for three weeks (60 mg of n-3 LCPUFA per

114

Chapter 4. Dietary modeling of n-3 LCPUFA

day) resulted in significant increases of plasma n-3 LCPUFA from 2.0 to 2.4 % and in phospholipids from 3.9 to 4.3 %(31). A 4-week double-blind cross-over study in healthy individuals aged over 45 years showed that consumption of one n-3 enriched egg containing 9.3 % ALA, 0.2 % EPA and 1.5 % DHA improved CVD risk markers, including an increase in apolipoprotein A1 (Apo-A1) and decrease in Apo-B/Apo-A1 ratio, with no adverse impacts on blood lipids and inflammatory markers(38). Apo-A1 is a major component of HDL that is responsible for carrying cholesterol from arteries, whilst Apo-B carrying lipoproteins like LDL are responsible for carrying cholesterol to tissues, and they are accepted as superior CVD risk markers(39). It has been reported that Apo-B/Apo-A1 ratio was associated with metabolic syndrome in obese children(40). Consumption of six n-3 fatty acid enriched eggs per week by healthy lacto-ovo- vegetarian adults increased serum lutein, an antioxidant that protects the macula from light-initiated oxidative damage(41), by 18 %(42). Similarly, the consumption of one egg enriched with DHA, vitamin E, selenium and lutein significantly increased plasma concentrations of DHA, alpha-tocopheroland lutein in adults(43). An average intake of 300 mg EPA plus DHA in milk resulted in a 25-50 % improvement in plasma levels of fatty acids after 6 weeks(27). It has reported that a yoghurt drink was the best matrix for rapid absorption of n-3 LCPUFA, which might be due to the preformed emulsions in the yogurt(44). In our Australian study, children who consumed milk showed the greatest increase in EPA and total n-3 LCPUFA intake after modelling, while children who consumed bread showed the greatest increase in DHA. Although the n-3 LCPUFA enriched egg in this study contained the highest level of DHA compared to bread, milk and yogurt, this amount did not contribute to the increase in DHA intake, as only 23 % of all children consumed eggs during the survey.

Uptake of foods enriched with n-3 LCPUFA by Australian consumers is low(45), with barriers including excessive price(46), undesirable sensory qualities (fishy after taste)(16) and a possibility of overdosing(45,47). Consumers want realistic advice(48) in terms of how to adopt a dietary recommendation on a daily basis. Our dietary modelling provides a realistic example of how changes to n-3 LCPUFA intakes can be achieved by consuming different combinations of n-3 LCPUFA enriched foods. This study demonstrates that in children who do not consume fish, promotion of foods enriched with n-3 LCPUFA may be a strategy to increase intakes, however the dietary recommendations of 500 mg/d for cardiovascular health(13) or SDT range of 300 to 610 115

Chapter 4. Dietary modeling of n-3 LCPUFA

mg/d(15) will still not be reached. A wider range of appropriate n-3 LCPUFA enriched foods would help children to meet the recommendation. Increasing the levels of n-3 LCPUFA added to enriched products is unlikely, due to cost, as well as the existence of Australian food regulations that allow permitted maximum levels within food categories. In our dietary modelling, replacement of bread, egg, milk and yogurt with the corresponding n-3 LCPUFA enriched products increased the average cost of consuming these foods by AUD 45 cent per day. It remains to be demonstrated whether enrichment as a public health strategy is an acceptable option for Australian families.

A potential limitation of this study is that the use of two 24-hour food recalls to obtain the n-3 LCPUFA intake may introduce errors in estimating habitual consumption of n-3 LCPUFA food sources, such as enriched foods and especially fish, which was only occasionally consumed. This may lead to potential under or over-reporting bias. However, compared to previous national dietary surveys in 1985 and 1995, as well as in Western Australia(29), the trend of low fish consumption which results in the low n-3 LCPUFA intakes appears to have remained unchanged in the population of Australian children. Similar findings have also been reported in a recent study of overweight and obese children(49) as well as in our previous survey in Australian families with young children(47).

Our estimation using adjusted data found lower intakes of EPA, DPA, DHA and total n-3 LCPUFA than average n-3 LCPUFA intake estimations using unadjusted data (Online Supplementary Table 1, 2A and 2B). This could be explained by “extreme” values of n-3 LCPUFA which may result in non-normal data which could be responsible for wide distributions of intakes and enlarged variances. Our analysis using the MSM showed an indication that the Box-Cox transformation algorithm did not find an optimal lambda. The MSM selected the “best” not optimal transformation parameter lambda and searched for lambda over a grid to minimize the skewness to account for intra- and inter-individual variations in usual intake distribution. Therefore, the distribution intake of this study should be carefully interpreted(22).

In conclusion, this dietary modelling scenario provides a realistic example of practical ways that Australian children can increase their n-3 LCPUFA intakes without major changes to dietary behavior by consuming products enriched with n-3LCPUFA. Consumption of bread, egg, milk and yogurt enriched with n-3 LCPUFA in this 116

Chapter 4. Dietary modeling of n-3 LCPUFA

scenario doubled the median population intake of n-3 LCPUFA and importantly increased median DHA intakes 7-fold. The increased DHA intakes were more pronounced in the non-fish eating group, as median DHA increased 23-fold. However, consumption of n-3 LCPUFA enriched foods does not increase intake sufficiently to achieve the suggested dietary targets for prevention of chronic disease (SDT). The best way to meet n-3 LCPUFA intakes for optimal health is consumption of fish, or fish oil supplements for non-fish consumers.

Acknowledgements

We would like to thank to the participants and the National Health and Medical Research Council (NHMRC) registered ethics committees of Commonwealth Scientific and Industrial Research Organization and University of South Australia, also the Australian Social Science Data Archive for permission using the 2007 ANCANPA survey dietary data. Dr Jimmy Louie, School of Health Sciences is thanked for his assistance with using the MSM. Sven Knüppel, MSc Epidemiology, Department of Epidemiology at German Institute of Human Nutrition Potsdam-Rehbrücke is thanked for his assistance in interpretation the results of the MSM. We would also like to thank the Directorate General of Higher Education Indonesia for sponsoring Setyaningrum Rahmawaty, a lecturer from the University of Muhammadiyah Surakarta Indonesia for her PhD at the University of Wollongong, New South Wales, Australia.

Conflicts of interest

No competing interests were identified.

117

Chapter 4. Dietary modeling of n-3 LCPUFA

Supplement Table 1. Modelling of the average intake of EPA, DPA, DHA and total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods (n = 4487)a EPA DHA Total n-3 LCPUFAb n-3 enriched foods Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Baseline (actual diet) 21.8 ± 54.7 7.2 (2.0-20.6) 39.0 ± 103 6.0 (0.9-26.7) 79.2 ± 173 28.9 (10.9-72.5) 1 item Model 1 (bread) 23.0 ± 54.7 8.5* (3.3-21.8) 44.9 ± 103 14.5* (5.4-35.0) 86.2 ± 173 36.3* (18.0-79.0) Model 2 (egg) 25.5 ± 55.7 11.1* (3.0-26.7) 44.3 ± 106 6.5* (0.9-37.5) 87.9 ± 177 32.5* (11.3-89.8) Model 3 (milk) 21.9 ± 54.7 7.3* (2.2-20.9) 44.9 ± 103 15.3* (3.9-36.8) 85.1 ± 173 34.9* (16.8-78.6) Model 4 (yogurt) 21.8 ± 54.7 7.2 (2.1-20.6) 43.1 ± 105 7.6* (1.2-37.4) 83.0 ± 174 32.3* (11.9-79.6) 2 items

Model 5 (bread, egg) 26.8 ± 55.7 12.3* (4.4-28.1) 50.0 ± 106 14.8* (5.7-44.1) 94.9 ± 177 41.0* (18.6-97.4) Model 6 (bread, milk) 23.2 ± 54.8 8.7* (3.5-22.0) 50.7 ± 103 21.4* (11.2-42.6) 92.0 ± 174 42.9* (23.6-86.1) Model 7 (bread, yogurt) 23.0 ± 54.7 8.6* (3.3-21.8) 48.8 ± 104 16.3* (6.1-43.3) 90.0 ± 174 40.4* (19.2-86.9) Model 8 (milk, yogurt) 21.9 ± 54.7 7.3* (2.2-20.9) 48.8 ± 104 18.1* (5.4-42.9) 88.9 ± 174 38.8* (18.5-85.2) Model 9 (milk, egg) 25.7 ± 55.7 11.2* (3.2-26.9) 50.0 ± 106 15.7* (4.0-44.2) 93.8 ± 177 39.5* (17.3-95.7) Model 10 (yogurt, egg) 25.5 ± 55.7 11.1* (3.0-26.7) 48.1 ± 107 8.1* (1.2-45.7) 91.8 ± 178 37.1* (12.2-95.7) 3 items Model 11 (bread, milk, yogurt) 23.2 ± 54.8 8.7* (3.5-22.0) 54.5 ± 104 23.7* (12.4-49.9) 95.9 ± 174 46.7* (25.2-93.0) Model 12 (bread, yogurt, egg) 26.8 ± 55.7 12.3* (4.4-28.1) 53.8 ± 107 17.0* (6.4-52.5) 98.8 ± 178 45.9* (19.7-104) Model 13 (milk, yogurt, egg) 25.7 ± 55.7 11.2* (3.2-26.9) 53.8 ± 107 18.8* (5.7-52.5) 97.7 ± 178 43.9* (18.8-101) Model 14 (bread, egg, milk) 26.9 ± 55.7 12.3* (4.5-28.3) 55.7 ± 107 22.2* (11.4-51.9) 100.8 ± 178 47.3* (24.0-104) 4 items Model 15 (bread, egg, milk, yogurt)c 26.9 ± 55.7 12.3* (4.5-28.2) 59.5 ± 107 25.4* (12.7-58.8) 104.6 ± 178 51.4* (25.7-109) aFrom two 24-hour dietary recalls EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; total n-3 LCPUFA, sum EPA+DPA+DHA; SD, standard deviation; IQR, Interquartile range; cThe average cost of meal in children who ate bread, egg, milk and yogurt increased AUD 45 cent per day *Significant difference between the baseline (actual intake) and intake from each model (p = 0.001) using Post-hoc analysis with Wilcoxon Signed-Rank Tests with a Bonferroni correction applied

118

Chapter 4. Dietary modeling of n-3 LCPUFA

Supplement Table 2A. Modelling of the average intake of EPA, DPA, DHA and total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foodsa (n = 3554)b in non-fish eaterc Non-fish eater (n = 3554) EPA DHA Total n-3 LCPUFA Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Baseline (actual diet) 9.2 ± 13.1 4.8 (1.3-11.9) 9.3 ± 15.7 3.2 (0.3-11.4) 33.0±41.0 20.8 (8.1-42.7) 1 item Model 1 (bread) 10.5 ± 13.2 6.2* (2.7-13.2) 15.2 ± 17.5 11.5* (3.5-2.0) 40.1 ± 42.3 28.1* (15.3-51.4) Model 2 (egg) 12.8 ± 16.0 7.1* (2.1-17.7) 14.0 ± 28.4 3.4* (0.3-12.9) 41.3 ± 53.5 22.1* (8.4-53.3) Model 3 (milk) 9.3 ± 13.1 5.0* (1.4-12.0) 15.0 ± 17.8 10.6* (2.4-21.4) 38.9 ± 42.2 27.0* (14.0-48.4) Model 4 (yogurt) 9.2 ± 13.1 4.8 (1.3-11.9) 12.9 ± 21.3 4.0* (0.6-17.9) 36.6 ± 43.6 23.3* (8.9-47.8) 2 items Model 5 (bread, egg) 14.0 ± 16.2 8.4* (3.4-18.9) 19.9 ± 29.4 11.5* (3.7-22.0) 48.4 ± 54.4 30.0* (15.8-61.2) Model 6 (bread, milk) 10.6 ± 13.2 6.4* (2.9-13.4) 20.9 ± 19.3 16.8* (8.6-28.3) 46.0 ± 43.4 34.1* (20.6-57.4) Model 7 (bread, yogurt) 10.5 ± 13.2 6.2* (2.7-13.2) 18.8 ± 22.6 12.4* (4.1-24.2) 43.7 ± 44.8 30.6* (16.2-57.2) Model 8 (milk, yogurt) 9.4 ± 13.1 5.0* (1.4-12.0) 18.7 ± 22.0 13.4* (3.2-24.9) 42.5 ± 44.2 29.8* (15.5-54.8) Model 9 (milk, egg) 12.9 ± 16.1 7.2* (2.2-17.8) 19.8 ± 29.5 10.9* (2.5-24.0) 47.2 ± 54.3 28.5* (14.3-59.2) Model 10 (yogurt, egg) 12.8 ± 16.1 7.1* (2.1-17.7) 17.7 ± 31.7 4.3* (0.6-21.4) 45.0 ± 55.4 25.3* (9.2-59.4) 3 items Model 11 (bread, milk, yogurt) 10.6 ± 13.2 6.4* (2.9-13.4) 24.5 ± 23.2 19.0* (10.3-32.7) 49.6 ± 45.4 37.1* (22.0-62.3) Model 12 (bread, yogurt, egg) 14.0 ± 16.2 8.4* (3.4-18.9) 23.5 ± 32.7 12.8* (4.3-29.2) 52.0 ± 56.4 33.2* (16.5-67.7) Model 13 (milk, yogurt, egg) 12.9 ± 16.1 11.0* (3.1-26.9) 23.4 ± 32.1 13.9* (3.3-29.6) 50.8 ± 55.8 32.0* (15.6-64.9) Model 14 (bread, egg, milk) 14.2 ± 16.2 8.6* (3.5-19.0) 25.7 ± 30.4 17.3* (8.8-31.3) 54.3 ± 55.2 36.3* (20.9-67.4) 4 items Model 15 (bread, egg, milk, yogurt) 14.2 ± 16.2 8.6* (3.5-19.0) 29.3 ± 33.0 19.5* (10.4-36.3) 57.9 ± 56.8 39.8* (22.5-73.6)

aModelled diet by replacing bread, egg, milk and yogurt with n-3 enriched for these foods in the actual diet of each child for all children (as note in Table 1) bFrom two 24-hour dietary recalls; cFish eater, ate fish during at least one of the two survey days; Non-fish eater, did not eat fish during the two survey days EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; total n-3 LCPUFA, sum EPA+DPA+DHA; SD, standard deviation; IQR, Interquartile range; *Significant difference between the baseline (actual intake) and intake from each model (p = 0.001) using Post-hoc analysis with Wilcoxon Signed- Rank Tests with a Bonferroni correction applied

119

Chapter 4. Dietary modeling of n-3 LCPUFA

Supplement Table 2B. Modelling of the average intake of EPA, DPA, DHA and total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foodsa (n = 933)b in fish eaterc Fish eater (n = 933) EPA DHA Total n-3 LCPUFA Mean ± SD Median (IQR) Mean ± Median (IQR) Mean ± SD Median (IQR) SD Baseline (actual diet) 69.7 ± 104 35.8 (19.0-71.5) 153 ± 182 87.0 (39.6-203) 255 ± 314 150 (76.0-308) 1 item Model 1 (bread) 70.8 ± 104 36.9* (20.0-72.5) 158 ± 183 94.0* (45.8-205) 262 ± 314 157* (83.0-312) Model 2 (egg) 74.1 ± 105 39.7* (21.5-79.2) 159 ± 185 96.0* (41.5-210) 266 ± 317 163* (81.7-319) Model 3 (milk) 69.8 ± 104 35.9* (19.1-71.7) 159 ± 182 93.0* (47.5-206) 261 ± 314 155* (83.0-311) Model 4 (yogurt) 69.7 ± 104 35.8 (19.0-71.5) 158 ± 182 96.0* (42.5-204) 260 ± 314 157* (80.3-311) 2 items Model 5 (bread, egg) 75.2 ± 105 41.1* (22.6-80.2) 164 ± 185 102* (47.7-214) 272 ± 317 167* (87.7-325) Model 6 (bread, milk) 70.9 ± 104 37.0* (20.4-72.9) 164 ± 183 98.0* (52.3-213) 267 ± 315 161* (88.4-318) Model 7 (bread, yogurt) 70.8 ± 104 36.9* (20.2-72.5) 163 ± 183 100* (48.4-209) 266 ± 315 165* (87.0-314) Model 8 (milk, yogurt) 69.8 ± 104 35.9* (19.1-71.7) 164 ± 183 100* (50.7-210) 266 ± 314 164* (87.3-316) Model 9 (milk, egg) 74.3 ± 105 39.8* (21.6-79.3) 165 ± 185 100* (49.1-218) 272 ± 317 168* (87.6-324) Model 10 (yogurt, egg) 74.1 ± 105 39.7* (21.5-79.2) 164 ± 185 106* (45.6-211) 271 ± 317 170* (85.7-322) 3 items Model 11 (bread, milk, yogurt) 70.9 ± 104 37.0* (20.4-72.9) 169 ± 183 106* (55.0-215) 272 ± 315 169* (92.6-320) Model 12 (bread, yogurt, egg) 75.2 ± 105 41.1* (22.6-80.2) 169 ± 185 109* (51.0-220) 277 ± 317 178* (91.3-329) Model 13 (milk, yogurt, egg) 74.3 ± 105 39.8* (21.6-79.3) 170 ± 185 108* (52.6-220) 276 ± 317 175* (91.2-324) Model 14 (bread, egg, milk) 75.4 ± 105 41.2* (22.7-80.3) 170 ± 185 106* (54.0-222) 278 ± 317 173* (93.8-329) 4 items Model 15 (bread, egg, milk, 75.4 ± 105 41.24* (22.4-80.3) 175 ± 185 113* (58.1-223) 283 ± 317 183* (97.7-332) yogurt) aModelled diet by replacing bread, egg, milk and yogurt with n-3 enriched for these foods in the actual diet of each child for all children (as note in Table 1) bFrom two 24-hour dietary recalls; cFish eater, ate fish during at least one of the two survey days EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; n-3 LCPUFA, omega-3 long chain polyunsaturated fatty acid; total n-3 LCPUFA, sum EPA+DPA+DHA; SD, standard deviation; IQR, Interquartile range; *Significant difference between the baseline (actual intake) and intake from each model (p = 0.001) using Post-hoc analysis with Wilcoxon Signed-Rank Tests with a Bonferroni correction applied

120

Chapter 4. Dietary modeling of n-3 LCPUFA

700 Non fish eaters (mean) Non fish eaters (median) Fish eaters (mean) Fish eaters (median) 600

500 range of SDT

3 LCPUFA (mg/d) 400 -

300

200

100 range of AI

0 Average Average intake totalof n M 1 M 2 M 3 M 4 M 5 M 6 M 7 M 8 M 9 M 10 M 11 M 12 M 13 M 14 M 15 Actual 1 items 2 items 3 items 4 items diet

Supplement Fig 1. Modelling intake of total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods from (n = 4487) in relation to the nutrient reference values(15)

SDT, suggested dietary target; AI, adequate intake; M1 to M15 refers to Models 1 to 15 for replacement with n-3 enriched foods (see details in Table 2)

121

Chapter 4. Dietary modeling of n-3 LCPUFA

Non fish eater (meet AI) Non fish eater (meet SDT) Fish eater (meet AI) Fish eater (meet SDT)

100

40 % ofchildren % 20

0 Actual M 1 M 2 M 3 M 4 M 5 M 6 M 7 M 8 M 9 M 10 M 11 M 12 M 13 M 14 M 15 diet 1 item 2 items 3 items 4 items

Supplement Fig 2. Modelling intake of total n-3 LCPUFA (non-adjusted intra individual variances) before and after replacement with n-3 enriched foods from (n = 4487) in relation to the nutrient reference values(15)

SDT, suggested dietary target; AI, adequate intake; M1 to M15 refers to Models 1 to 15 for replacement with n-3 enriched foods (see details in Table 2)

122

Chapter 4. Dietary modeling of n-3 LCPUFA

References

1 Australian Bureau of Statistics. Causes of Death, Australia, 2010. Canberra: Australian Bureau of Statistics, 2012. (ABS Cat. No. 3303.0). http://www.abs.gov.au/ausstats/[email protected]/Products/BBC4B00DFF0E942ACA2579C 6000F6B15?opendocument#: accessed on April, 23, 2012.

2 Australian Institute of Health and Welfare. Living dangerously: Australians with multiple risk factors for cardiovascular disease. Canberra: Australian Institute of Health and Welfare, 2005. (AIHW Cat. No. AUS 57.)

3 Mietus-Snyder M, Krauss RM. Lipid metabolism in children and adolescents: impact on vascular biology. J Clin Lipidol 2008; 2: 127-137.

4 Russo GL. Dietary n-6 and n-3 polyunsaturated fatty acids: from biochemistry to clinical implication for cardiovascular prevention. Biochem Pharm 2009; 77: 937- 946.

5 Riediger ND, Othman RA, Suh M et al. A systemic review of the roles of n-3 fatty acids in health and disease. J Am Diet Assoc 2009; 109: 668-679.

6 Romeo J, Wärnberg J, García-Mármol E et al. Daily consumption of milk enriched with fish oil, oleic acid, minerals and vitamins reduces cell adhesion molecules in healthy children. Nutr Metab Cardiovasc Dis 2011; 21: 113-120.

7 Dangardt F, Osika W, Chen Y et al. Omega-3 fatty acid supplementation improves vascular function and reduces inflammation in obese adolescents. Atherosclerosis 2010; 212: 580-515.

8 Engler MM, Engler MB, Malloy M, et al. Docosahexaenoic acid restores endothelial function in children with hyperlipidemia: results from the EARLY Study. Int J Clin Pharmacol Ther 2004; 42: 672–679.

9 Brenna JT, Salem N Jr, Sinclair AJ et al. Alpha-linolenic acid supplementation and conversion to n-3 long-chain polyunsaturated fatty acids in humans. Prostaglandins Leukot Essent Fatty Acids 2009; 80: 85-91.

10 Meyer BJ, Mann NJ, Lewis JL et al. Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids. Lipids 2003; 38: 391-398.

11 Whelan J, Rust C Innovative dietary sources of n-3 fatty acids. Annu Rev Nutr 2006; 26: 75-103.

12 Colquhoun D, Ferreira-Jardim A, Udell T et al. Review of evidence fish, fish oils, n- 3 pollyunsaturated fatty acids and cardiovascular health. 2008. Available at: www.heartfoundation.org.au: accessed on January, 15, 2011.

13 National Health Foundation of Australia. Position statement, fish, fish oils, n-3 polyunsaturated fatty acids and cardiovascular health. 2008. Available at:

123

Chapter 4. Dietary modeling of n-3 LCPUFA

http://www.heartfoundation.org.au/sitecollectiondocuments/fish-position- statement.pdf : accessed on January, 15, 2011.

14 National Health Medical Research Council. Nutrient reference value for Australia and New Zealand including recommended dietary intakes. Australian Government Department of Health and Ageing. Commonwealth of Australia 2006.

15 Meyer BJ, Kolanu N. Australian children are not consuming enough long-chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 2011; 27: 1136- 1140.

16 Patch CS, Tapsell LC, Mori TA et al. The use of novel foods enriched with long chain n-3 fatty acids to increase dietary intake: a comparison of methodologies assessing nutrient intake. J Am Diet Assoc 2005: 105: 1918-1926.

17 Lovegrove JA, Brooks CN, Murphy MC et al. Use of manufactured foods enriched with fish oils as a means of increasing long chain n-3 polyunsaturated fatty acid intake. Br J Nutr 1997; 78: 223-236.

18 Murphy KJ, Meyer BJ, Mori TA et al. Impact of food enriched with n-3 long-chain polyunsaturated fatty acids on erythrocyte n-3 levels and cardiovascular risk factors. Br J Nutr 2007; 97: 749-757.

19 Commonwealth of Australia. Main Findings - 2007 Australian National Children’s Nutrition and Physical Activity Survey. 2008. Available at: http://www.health.gov.au/internet/main/publishing.nsf/content/66596E8FC68FD1A3 CA2574D50027DB86/$File/childrens-nut-phys-survey.pdf: accessed on June, 23, 2010.

20 Australia Social Science Data Archive. 2009. Available at: http://assdanesstar.anu.edu.au/webview/?object = http://assda-nesstar.anu.edu.au /obj/fCatalog/Catalog28: accessed on January, 7, 2009.

21 Department of Health and Ageing. User guide - 2007 Australian National Children’s Nutrition and Physical Activity Survey. 2007. Available at: http://www.health.gov.au/internet/main/publishing.nsf/Content/phd-nutrition- childrens-survey-userguide: accessed on June, 23, 2010.

22 EFCOVAL. Multiple Source Method (MSM) for estimating usual dietary intake from short-term measurement data – User guide. 2011. Available at: https://msm.dife.de/: accessed on March, 27, 2013.

23 Allen L, Benoist B, Dary O et al. Guidelines on food fortification with micronutrients. World Health Organisation, Food and Agricultural Organization of the United Nations. 2006. Available at: http://www.who.int/nutrition/publications/guide_food_fortification_micronutrients.p df: accessed on April, 10, 2012.

24 Garg ML, Wood LG, Singh H et al. Means of delivering recommended levels of long chain n-3 polyunsaturated fatty acids in human diets. J Food Sci 2006; 71: R66-R71.

124

Chapter 4. Dietary modeling of n-3 LCPUFA

25 Australian Government, Department of Health and Ageing, National Health and Medical Research Council. Food for health, dietary guidelines for Australian, a guide to healthy eating. 2005. Available at: http://www.nhmrc.gov.au/_files_nhmrc /publications/attachments/ n31.pdf?q=publications/synopses/_files/n31.pdf: accessed on February, 12, 2012.

26 Mantzioris E, Cleland LG, Gibson RA et al. Biochemical effects of a diet containing foods enriched with n-3 fatty acids. Am J Clin Nutr 2000; 72: 42-48.

27 Lopez-Huertas E. Health effects of oleic acid and long chain omega-3 fatty acids (EPA and DHA) enriched milks. A review of intervention studies. Pharmacological Res 2010; 61: 200-207.

28 Harris WS. International recommendations for consumption of long-chain omega-3 fatty acids. J Cardio Med 2007; 8(suppl.1): S50-S52.

29 Clayton EH, Hanstock TL, Watson JF. Estimated intakes of meat and fish by children and adolescents in Australia and comparison with recommendations. Br J Nutr 2009; 101: 1731-1735.

30 Howe P, Meyer B, Record S et al. Dietary intake of long-chain ω-3 polyunsaturated fatty acids: contribution of meat sources. Nutrition 2006; 22: 47-53.

31 Yep YL, Li D, Mann NJ et al. Bread enriched with microencapsulated tuna oil increases plasma docosahexaenaoic acid and total omega-3 fatty acids in humans. Asia Pac J Clin Nutr 2002; 11: 285-291.

32 Harris WS, Von Schacky C. The omega-3 index: a new risk factor death from coronary heart disease? Prev Med 2004; 39: 212-220.

33 Ferrier LK, LJ Caston, S Leeson et al. α-Linolenic acid-and docosahexaenoic acid- enriched eggs from hens fed flaxseed: influence on blood lipids and platelet phospholipid fatty acids in humans. Am J Clin Nutr 1995; 62: 81-86.

34 Jian Z, Sim JS. Consumption of n-3 polyunsaturated fatty acis-enriched eggs and changes in plasma lipids of human subjects. Nutrition 1993; 9: 513-518.

35 Delaroudis S, Slavakis A, Kyroudi A et alOmega-3 PUFA modified eggs: a study on their effect on the lipid profile of human volunteers. 71st EAS’ Congress and Satellite Symposia 1999; 130-131.

36 Lewis NM, K Schalch, Scheideler. Serum lipid response to n-3 fatty acid enriched eggs in persons with hypercholesterolemia. J Am Diet Assoc 2000; 100: 365-367.

37 Mu H. Bioavailability of omega-3 long-chain polyunsaturated fatty acids from foods. Agro Food Industry hi-tech 2008; 19: 24-26.

38 Öhman M, Åkerfeldt T, Nilsson I et al. Biochemical effects of consumption of eggs containing omega-3 polyunsaturated fatty acids. Upsala J Med Sci 2008; 113: 315- 324.

125

Chapter 4. Dietary modeling of n-3 LCPUFA

39 Walldius G, Jungner I. Apolipoprotein B and apolipoprotein A-I: risk indicators of coronary heart disease and targets for lipid-modifying therapy. J Intern Med 2004; 255: 188-205.

40 Savas Erdeve S, Simsek E, Dallar Y et al. Utility of ApoB/ApoA1 ratio for the prediction of cardiovascular risk in children with metabolic syndrome. Indian J Pediatr 2010; 77: 1261-1265.

41 Nolan JM, Stack J, O’Connell E et al. The relation between macular pigment optical density and its constituent carotenoids in diet and serum. Invest Ophthalmol Vis Sci 2007; 48: 571-582.

42 Burns-Whitmore BL, Haddad EH, Sabate J et al. Effect of n-3 fatty acid enriched eggs and organic eggs on serum lutein in free-living lacto-ovo vegetarians. Eur J Clin Nutr 2010; 64: 1332-1337.

43 Surai P, MacPherson A, Speake B et al. Designer egg evaluation in a controlled trial. Eur J Clin Nutr 2000; 54: 298-305.

44 Schram LB, Nielsen CJ, Porsgaard T et al. Food matrices affect the bioavailability of (n-3) polyunsaturated fatty acids in a single meal study in humans. Food Res Inter 2007; 40: 1062-1068.

45 Cox DN, Evans G, Lease HJ. Predictors of Australian consumers’ intentions to consume conventional and novel sources of long-chain omega-3 fatty acids. Public Health Nutr 2007; 11: 8-16.

46 Patch CS, Tapsell LC, Williams PG. Overweight consumers’ salient beliefs on omega-3-enriched functional foods in Australia’s Illawarra region. J Nutr Educ Behav 2005; 37: 83-89.

47 Rahmawaty S, Charlton K, Lyons-Wall P et al. Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Nutr Diet 2013; DOI: 10.1111/1747-0080.12022

48 International Food Information Council Foundation (2007) Fitting dietary fats into a healthful diet: A consumer point of view. Available at: http://www.foodinsight.org/Resources/Detail.aspx?topic=Fitting_Dietary_Fats_into _a_Healthful_Diet_A_Consumer_Point_of_View_ : accessed on March, 27, 2012.

49 Burrows T, Berthon B, Garg ML et al. A comparative validation of a child food frequency questionnaire using red blood cell membrane fatty acids. Eur J Clin Nutr 2012; 66: 852-829.

126

Chapter 5. Food pattern study

Chapter 5

The majority of this section is the substantive content of the work published in Nutrition 2013, http://dx.doi.org/10.1016/j.nut.2013.07.014

Statement

As the primary supervisor, I, Barbara J Meyer, declare that the greater part of the work in this paper is attributed to the candidate, Setyaningrum Rahmawaty.

In the manuscript, Setyaningrum contributed to study design and was primarily responsible for data collection, data analysis, data interpretation and writing up the manuscript.

A/Prof Barbara J Meyer (Main Supervisor)

127

Chapter 5. Food pattern study

Food patterns of Australian children aged 9-13 years in relation to omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) intake

(Nutrition 2013, http://dx.doi.org/10.1016/j.nut.2013.07.014)

Setyaningrum Rahmawatya,b, Philippa Lyons-Wallc, Marijka Batterhamd, Karen Charltonb, Barbara J Meyera,b

Corresponding author. Tel.: +61 (0)2 4221 3459, Fax: +61 (0)2 4221 5945 Email address: [email protected] (B.J. Meyer)

SR performed the statistical analysis and interpretation of the data and drafted the manuscript, PLW participated in study design and assisted with the first draft of the manuscript, MB provided statistical consultation, KC and BJ co-authored the manuscript and were involved in conceptualization of the study. All authors contributed to the manuscript, read and approved the final manuscript. No conflicts of interest were identified by the authors.

128

Chapter 5. Food pattern study

Abstract

Objective: This study examined foods patterns of Australian children aged 9-13 years in relation to omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) intake.

Research Methods and Procedures: Secondary analysis was conducted on nationally representative food data of 1,110 Australian children aged 9-13 years (525 boys, 585 girls) that was obtained using two 24-hour recalls. Principle component factor analysis was used to identify food patterns. Discriminant function analysis was used to identify the relationship between the food patterns and total n-3 LCPUFA intake.

Results: Four major food patterns emerged in each sex. For boys these were labelled: ‘snack foods’, ‘soft drinks’, ‘vegetables’ and ‘pork and meat chops, steak and mince’, and for girls: ‘vegetables’, ‘take aways’, ‘tea, coffee, iced coffee drinks’ and ‘canned meals and soup’. Fish consumption bought from take-away outlets was more frequently consumed in the ‘soft drink’ (r = 0.577) and ‘take aways’ (r = 0.485) food pattern in boys and girls, respectively. In contrast, fish consumption prepared at home was more often consumed in ‘vegetables’ in both boys (r = 0.018) and girls (r = 0.106), as well as in the ‘pork and meat chops, steak and mince’ food pattern in boys (r = 0.060). There was a trend that in boys, the ‘vegetables’ group discriminated children who consumed n-3 LCPUFA levels similar to adequate intakes (AI) (p = 0.067), while in girls, the ‘take aways’ food pattern discriminated for being a fish consumer (p = 0.060).

Conclusion: Dietary patterns associated with a high consumption of vegetables in boys and ‘take aways’ that include meat and fish in girls are likely to positively influence dietary n-3 LCPUFA intake in Australian children.

Keywords: dietary patterns, n-3 LCPUFA, children, Australia

Introduction

Sufficient dietary intake of omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) is required for optimal health, to reduce the incidence of chronic diseases such as cardiovascular diseases (CVD) in adults [1, 2] and to prevent obesity-related chronic diseases in children [3]. In Australia, the National Health and Medical 129

Chapter 5. Food pattern study

Research Council (NHMRC) has established the nutrient reference value for n-3 LCPUFA in children younger than 14 years based on adequate intake (AI) data [4] or observed median intakes from the National Dietary Survey[5]. For children aged 14-16 years, a suggested dietary target (SDT) recommended for prevention of chronic diseases, has been set at 610 and 430 mg/d for boys and girls, respectively, based on the observed 90th percentile of the population intake [4]. The AI is defined as ‘the average daily nutrient intake level based on observed or experimentally-determined approximations or estimates of nutrient intake by a group (or groups) of apparently healthy people that are assumed to be adequate’ and the SDT is defined as ‘a daily average intake from food and beverages for certain nutrients that may help in prevention of chronic disease’ [4]. Hence the AI merely reflects the median intakes of the population and is not a recommended intake, while the SDT is a target intake for optimal health. The National Heart Foundation of Australia recommends an n-3 LCPUFA intake of 500 mg per day for adults and that children should follow the adult recommendation [6]. Meyer and Kolanu (2011) have extrapolated SDTs for children younger than 14 years from adjusted energy intakes, by sex and age group [7].

The contribution of fish, particularly oily fish, as well as foods enriched with n-3 LCPUFA to improving n-3 LCPUFA intake has been widely reported [8, 9] including in Australian children [10]. However, only approximately a fifth of Australian children consume fish regularly and less than 7% of children include n-3 enriched foods in their diets [7]. Our group has previously reported, using dietary modeling, that substitution of Australian children’s intakes of bread, egg, milk and yoghurt with n-3 enriched products that are commercially available in supermarkets would significantly increase n-3 LCPUFA intake [11], but not to the suggested dietary target level recommended by the NHMRC [4].

Individuals consume meals or dishes that commonly comprise a variety of foods with complex combinations of nutrients that are likely to be interactive or synergistic [12], therefore it is necessary to examine the effect of overall diet rather than a single food when evaluating intakes of specific nutrients in relation to health outcomes [13]. Food pattern analysis has been used to examine the relationship between overall dietary patterns and the risk of chronic diseases [14,15], nutritional status [16,17]. This method has also been used to evaluate dietary guidelines or dietary recommendations due to the

130

Chapter 5. Food pattern study

capability of the food pattern analysis to generate a broader picture of food consumption in a population [13]. It has been recommended that differences in dietary patterns between gender should be examined specifically when assessing the influence of parental lifestyle on the children’s food preferences [18,19] like fish consumption. Moreover, differences in disease risks such as coronary heart diseases were exist between sexes [20], and it can be reduced by consuming fish [21]. The aim of this study was to explore food patterns of Australian children in relation to n-3 LCPUFA intake, in the context of the whole diet.

Materials and Methods

Subjects

Food data from a nationally representative sample of 1,110 children aged 9-13 years (525 boys and 585 girls), obtained from the 2007 Australian National Children’s Nutrition and Physical Activity Survey (Children’s survey) [22], were included in this study. The survey was conducted between February and August 2007 according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the NHMRC registered ethics committees of the Commonwealth Scientific and Industrial Research Organization and the University of South Australia [22]. Permission to access the dataset was obtained from the Australian Social Science Data Archive [23].

Dietary assessment, food grouping and food pattern analysis

Dietary intake was assessed from two 24-hour recalls using a standardized multiple pass 24-hour dietary recall methodology during computer assisted personal interview (CAPI) and computer assisted telephone interview (CATI) [24].

A total of 500 food items in the two 24-hour recalls were grouped into 57 food groups based on the similarity of macronutrient composition (e.g. fat and fibre content), food behaviours (e.g. take- away, ethnic dishes and specialty items), mixed dishes, food sources of n-3 LCPUFA and foods that have the potential to be enriched with n-3 LCPUFA (Table 1). To examine food patterns associated with n-3 LCPUFA intake, intake of foods in each food group, rather than intake of specific nutrients, were used as principal components for analysis. Foods that did not contain n-3 LCPUFA’s were also 131

Chapter 5. Food pattern study

included in the modeling, for example sugar products and dishes, fruit products and dishes, confectionery and cereal/nut/fruit/seed bars, non-alcoholic beverages, dairy substitutes, seed and nut products and dishes, as well as legume and pulse products and dishes [10]. The percentage contribution from each food group for each respondent was then entered for components analysis within the factor analysis.

To investigate the underlying structure of the 57 food group items, the data were subjected to principal axis factoring with varimax rotation by gender. Prior to conducting the principal axis factoring, the data were examined for normality. Furthermore, a linear relationship was identified among the variables. The anti-image matrices were used to examine the suitability of the data for factor analysis. Twenty five food groups in boys and thirteen food groups in girls were excluded before running the factor analysis, because the Kaiser-Meyer-Olkin (KMO) value was below 0.5 and therefore the amount of variance in the data that could be explained by the factors was deemed unacceptable [25]. Four factors (with Eigenvalues or the variance of the factors exceeding 1) were identified in boys and in girls as underlying the thirty two and forty four food groups respectively (see Table 2). These factors (food patterns) were labeled based on the highest factor score in each factor.

Statistical analysis

The statistical analysis was carried out using Statistical Package for the Social Sciences (SPSS) software version 17.0, Chicago IL, USA. Intakes of eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and total n-3 LCPUFA (mg/d) are presented as mean ± SD and median (IQR). Intakes of n-3 LCPUFA were determined by fish eater status, based on consumption of at least one serve of fish (yes/no) from the two-24 hour recalls. Data were tested for normality using the Kolmogorov-Smirnov test and as the data were skewed, values were log-transformed (log10). Factor analysis was used to identify food patterns based on the 57 food groups. Discriminant function analysis was used to identify the relationship between food groups and the n-3 LCPUFA intake status. The scores obtained by the factor analysis were considered as continuous independent variables, and n-3 intake was considered as a dichotomous dependent variable [26], expressed as either achieving or not achieving nutrient reference values (AI and SDT), as well as by fish eater status. Separate regression analyses were

132

Chapter 5. Food pattern study

performed for each factor to test whether food patterns predicted changes in n-3 LCPUFA intake.

Results

Food patterns produced by factor analysis

Four food patterns were identified in boys and in girls, and were labeled according to the highest positive factor loading (Table 2). In boys, the identified food patterns ‘snack foods’, ‘soft drinks’, ‘vegetables’ and ‘pork and meat chops, steak and mince’ explained 5.3%, 5.2%, 5.2% and 5.1%, respectively, of the variance of food. In girls, derived food patterns were named ‘vegetables’, ‘take aways’, ‘tea, coffee, iced coffee drinks’ and ‘canned meals and soups’, which explained 4.2%, 4.1%, 3.6% and 3.6%, respectively, of the variance of food. In total, these factors accounted for 20.8% and 15.5% of the variance of food among boys and girls respectively (Table 2).

Table 1. Food grouping used in this study of Australian children (age = 9-13 years, n = 1,110) Food/ Food Groups Food items on dietary recall using CAPI and CATI 1. Chicken†b Chicken, turkey, chicken nuggets, kangaroo, nuggets, chicken roll, duck, lemon chicken, quail 2. Pork and meat chops, Pork, bacon, meat, ham, sweet and sour pork, savoury mince steak and mince† 3. Sausages†a Sausage, sausage roll, salami, cabanossi, devon, kabana, saveloy, pepperoni 4. Beef†a,b Beef, corned beef, veal, steak, pastrami, silverside, beef jerky, roast, venison 5. Lamb† Lamb 6. Take away (meat, fish†, Frankfurter, hotdog, McDonald, Kentucky Fried Chicken, meat cereal and vegetable sticks, patty, burger, burger rings, hamburger, pizza, Dominos, mixed) Burger King, Hungry jacks, pizza pocket, pizza hut, bakers delight product, Subway, pastie, meat pie, pie, Chiko roll, fried/deep fried fish bought from take away outlet and restaurant (squid or calamari, fish cake and fish finger crumbled) 7. Mixed meat dishes Curry, luncheon meat, mince meat, meatloaf, meatballs, curry puff, chilli con carne, casserole, stew, rissoles 8. Specialty (meat, cereal Taco, satay, dumpling, fajita, parmigiana, pad thai, bok choy, and vegetable mixed)a moussaka, dim sims, stroganoff, toad in the hole, souvlaki, chop suey, kebbeh, kebab, doner kebab, yiros, burrito, international food, spinach and feta roll, nachos, spring roll, recipe mean 9. Bread*a Toast, bread, bread roll, garlic bread, pita bread, French toast, mountain bread, fairy bread, roti 10. Sandwich* Sandwich, filled bread/roll 11. Muffina Muffin, English muffin

133

Chapter 5. Food pattern study

12. Savoury biscuit & other Bagel, savoury biscuits, wrap*, foccacia, tortilla, rice craker, bread typea pide, pappadam, Corn Thins, Jatz, Damper, Nashu, crumpets 13. Specialty cereal Sushi, couscous, polenta, idli (mixed)a,b 14. Noodlesa Noodles, chow mein, laksa, mie goreng 15. Breakfast cerealsb Cereal*, all bran, Weetbix , oats, Up and Go, barley, Quinoa, Wheat, Pupped wheat, porridge 16. Pastaa Pasta, lasagne, canned spaghetti, pasta sauce, macaroni cheese, pasta bake, spaghetti, homemade pasta, Bolognese sauce, bolognase, ravioli, cannelloni, carbonara 17. Rice Rice, risotto, fried rice, infant food 18. Sweet biscuit Sweet biscuit 19. Cake Cake, rice cake, tart, cup cake, custard tart 20. Pastry Pastry, pie, pancakes, bun (sweet), lamingtons, croissant, pikelets, doughnut, Loaf-sweet, hot cross bun, donut, cone/wafer, apple pie, filo pastry, Tim Tam, scones, strudel, brownie, profiterole, halawa, crepe, vol au vonts, souffle, crumble, roll, chiko roll, waffles, danish, snow delight 21. Dessert & puddingsa Pudding, Chocolate mousse, Mousse 22. Milk*a Milk 23. Flavoured milkb Milo, milky way, chocolate milk, Sustagen (ready made), milkshake, flavoured milk, Milo powder, milo drink, Sustagen powder, soy milk, strawberry milk, smoothie, Sip ahh straw, Ovaltine, Toodler formula (powder), Nesquick, hot chocolate, hot cocoa, Nesquick powder, drinking chocolate, Cocoa (eg. bournville cocoa), Babbycino, Quick 24. Cream*a Cream, cream bun, sour cream, custard, bavarian cream 25. Yoghurt*a Yoghurt, frozen yoghurt, soy yoghurt 26. Cheese* Cheese, cheese cake 27. Juicea,b Juice, cordial syrup, Ribena 28. Fruit juicea Apple juice, juice concentrate, Boost juice, Punch, Nectar 29. tea, coffee, iced Gloria Jeans drink, tea, coffee, iced coffee coffee drinks 30. Soft drink Soft drink, Coke, Fanta, frozen soft drink, squash, lemonade, Slushy (frozen soft drink) 31. Specialty drink Energy drink, sports drink, protein drink, malt drink, Powerade, Grapetiser, Gatorade, Cider, Yakult, powdered beverage 32. Potato Potato, potatoes, potato salad, gnocchi, mash, hash browns 33. Sweet potato Sweet potato 34. Vegetables Salad, zucchini, corn, mushroom, pumpkin, broccoli, carrot, tomato, vegetable, cabbage, cucumber, celery, cauliflower, lettuce, spinach, asparagus, salad roll, mixed vegetables, green beans, beetroot, snow pea, sprout, dressing, leek, pickle, gherkin, brussels sprout, parsnip, silverbeet, swede, turnip, okra, seaweed, capsicum 35. Mixed vegetable dishesa Stir-fry, mashed vegetables, fritters, coleslaw, Asian greens, samosa 36. Confectionarya Chocolate, lollies, Carob (chocolate), lollipop, Flake, Aero, Marshmallow, Nutella, M & MS, chewing gum, bubblegum, Twix, Fudge, Popslice, Kit Kat, Malteres, Rocky road, Crunchie, Time Out, licorice, chocolate crackles, Easter egg, Slice/balls (sweet), Ovalteenies, Mars bar, chocolate bar, Cherry ripe, Icy pole, Zooper doper, ice block

134

Chapter 5. Food pattern study

37. Snack bar†† Snack bar, fruit bar (e.g. Bellis), muesli bar, muffin bar, fruit bar, sport bar, snack bar, LCM, protein bar, Cheezels, cheese rings, picnic (bar) 38. Apple Apple 39. Bananaa Banana 40. Mandarin Mandarin 41. Other fruit Olive, grapes, kiwifruit, plum, orange, avocado, strawberry, rockmelon, pear, fruit, watermelon, mango, grape, apricot, raspberry, passionfruit, pineapple, peach, peaches, grapefruit, blueberry, pears, date, lemon, cantaloupe, apricots, melon, cherry, sultanas, cranberry, persimmon, prune, raisin, currant, figs, goji berries, tangerine, pomegranate, pawpaw, tangelo, guava, nectarine, honeydew melon 42. Fruit and dishesb Fruit salad, fruit snack, apple crumble, toffee apple, canned fruit, dried fruit, fruit crumble, sorbet, fruit leather/strap, rifle, fruche/frousse 43. Snack food Crisps, snack, crackers, dairy rice snack, Le snack, Cherrios, Oreo wafer stick, Twisties, Nachos, spring roll, rice crakers, roll up, pop corn, pretzel, pringels 44. Butter/margarine*a,b Butter, fat, margarine 45. Spreads*a,b Vegemite, spread, hummus, savoury, dip, peanut butter 46. Saucesa,b Gravy, mornay 47. Fish† Fish, salmon, tuna, fish fingers (includes canned fish) prepared from home 48. Other seafoodb Squid, lobster, scallop, prawns, calamari, seafood, prawn cracker, prawn toast, quiche 49. Egg†* Egg, omelette, scrambled egg 50. Legumes†† Beans, baked beans, lentils, tofu, kidney beans, falafels, bean curd, TVP, dhal 51. Nut and seed†† Nuts, peanuts, peas, almond, seeds, chick peas, homus, pistachio, almonds 52. Sugar product and Jelly, sugar, jam, honey, syrup, icing, topping, jelly beans, dishesa candyfloss, mint, Chupa chup, pavlova, Meringue, hundreds and thousands, sprinkles 53. Ice cream Ice cream, Thick shake, Cone 54. Canned meals and Soup, canned meal (does not include canned fish) soupa 55. Dietary supplementa,b Dietary supplement, fish oil, vitamin supplement, vitamin C, meal replacement powder, formula 56. Savoury saucesa Coconut milk, sauce simmer, herbs, onion, curry powder, coconut, garlic, chilli, ginger, tomato sauce, mustard, sauce, stock, tomato puree 57. Wine, beer & alcohola,b Alcoholic soda, Vodka, Beer, Ginger beer

†Natural sources of n-3 LCPUFA; †† Natural sources of n-3 ALA; * Products available in supermarket contained omega-3 (ALA or n-3 LCPUFA) aWas dropped for running in factor analysis in boys, bWas dropped for running in factor analysis in girls because the Kaiser-Meyer-Olkin (KMO) value bellowed 0.5 (no strong correlation)

In boys, the ‘snack foods’ pattern was characterized by high consumption (high positive factor loadings) of processed snack foods, chicken (all types, including nuggets,

135

Chapter 5. Food pattern study

curry), sandwiches and breakfast cereals, with low intakes (negative factor loadings) of fish, eggs and fatty meats; ‘soft drinks’ was characterized by high consumption of soft drink and take away items (meat, fish, cereal and vegetable mixed), with low consumption of fish; ‘vegetables’ was characterized by high intakes of vegetables, nuts and seeds and sweet potato; while ‘pork and meat chops, steak and mince’ was characterized by high consumption of pork and meat chops, steak and mince, cheese, potato, ice cream and soft drink with low consumption of fish and rare intake of egg.

In girls, the ‘vegetables’ pattern was characterized by high consumption of vegetables, potato, sausage, nuts and seeds, sweet potato and soft drink, with low consumption of fish and rare intake of yoghurt; ‘take-aways’ was characterized by high consumption of take-aways (meat, fish, cereal and vegetable mixed), speciality (meat, cereal and vegetable mixed), soft drink and snack food, with low consumptions of fish and eggs; ‘tea, coffee, iced coffee drinks ‘ was characterized by high consumption of tea, coffee, iced coffee drinks and mixed vegetable dishes, with low intakes of fish and yoghurt; while ‘canned meals and soups’ was characterized by high consumption of canned meals and soup and speciality drinks (such as energy drink, sports drink, protein drink, malt drink, Powerade, Grapetiser, Gatorade, Cider, Yakult, powdered beverage), with low consumption of yoghurt and egg (Table 2).

Characteristics of the children in each food pattern group, by gender are shown in Table 3. The characteristics of food patterns in this study were obtained from factor scores generated from the factor analysis, where this method produced continuous dietary factor scores for each child in each of the food pattern. High consumption of ‘take aways’ food including take away fish was shown in ‘soft drink’ and ‘take aways’ food patterns in boys and girls respectively (Table 3).

Children in the ‘soft drink’ food pattern in boys (r = 0.577) and ‘take-away’ food pattern in girls (r = 0.485) were consuming fish purchased from take-away outlets (including fish and chips, fish cakes, and fish fingers crumbed, fried or deep fried), more frequently than children in other food pattern groups, while intake of fish prepared from home and canned fish was relatively rare (r = -0.144 in boys, r = -0.046 in girls). Consumption of fish prepared from home and canned fish was slightly more frequent in the ‘vegetables’ food pattern in boys (r = 0.018) and girls (r = 0.106), and in the ‘pork and meat chops, steak and mince’ group in boys (r = 0.060) (Table 2). 136

Chapter 5. Food pattern study

Table 2. Food patterns a of Australian children by gender (age = 9-13 years, n = 1,110) Boysb Snack food Soft drink Vegetables Pork and meat chops, steak and mince Snack foodc 0.511d Soft drink 0.619 Vegetables 0.663 Pork and meat chops, steak and 0.764 Chicken† 0.493 Take aways (meat, fish†e, 0.577 Nut and seed†† 0.657 mince† Sandwich* 0.477 cereal and vegetable mixed) Sweet potato 0.594 Cheese* 0.758 Breakfast cereals 0.446 Sweet biscuit 0.293 Potato 0.286 Potato 0.363 Tea, coffee 0.348 Specialty drink 0.253 Chicken† 0.213 Ice cream 0.370 Ice cream 0.268 Egg†* 0.097 Lamb 0.177 Soft drink 0.185 Nut and seed†† 0.253 Fish† -0.144 Egg†* 0.074 Fish† 0.060 Rice 0.235 Cake -0.263 Fish† 0.018 Egg†* -0.120 Flavoured milk 0.226 Apple -0.363 Fish† -0.004 Other fruit -0.514 Egg†* -0.100 Pork and meat chops, steak -0.201 and mince† Girlsb Vegetables Take aways Tea, coffee, iced coffee drinks Canned meals and soup Vegetables 0.620 Take aways (meat, fish†e, 0.485 Tea and coffee 0.751 Soup 0.406 Potato 0.584 cereal and vegetable mixed) Mixed vegetable dishes 0.669 Specialty drink 0.353 Sausages† 0.434 Specialty meat, cereal and 0.431 Dessert and pudding 0.292 Snack bar†† 0.297 Nut and seed 0.344 vegetable mixed Pastry 0.233 Snack food 0.269 Sweet potato 0.338 Soft drink 0.442 Fatty meat 0.231 Cake 0.262 Soft drink 0.323 Snack food 0.414 Rice 0.221 Potato 0.217 Lamb† 0.249 Confectionary 0.289 Fish† -0.098 Pork and meat chops, steak and 0.215 Legumes†† 0.237 Noodles 0.207 Yoghurt* -0.099 mince† Rice 0.220 Fish† -0.046 Egg†* -0.152 Specialty meat, cereal and 0.220 Savoury sauces and 0.216 Egg†* -0.094 Sandwich* -0.202 vegetable mixed condiments Vegetables -0.228 Yoghurt* -0.080 Cake 0.210 Banana -0.280 Egg†* -0.114 Fish† 0.106 Other fruit -0.321 Mandarin -0.302 Egg†* 0.075 Apple -0.318 Fish† -0.490 -0.051 -0.421 -0.538 Yoghurt* Yoghurt* Bread* a Obtained from factor analysis with extraction method = principle component analysis, rotation method = varimax with Kaiser normalization and rotation converged in 15 iterations; b Percent variance explained for boys: snack food = 5.3%, soft drink = 5.2%, vegetables = 5.2%, fatty meat = 5.1%; girls: vegetables = 4.2%, take away = 4.1%, ake away = 3.6%, soup = 3.6%; c Food/food groups refer to Table 1; d Correlation coefficient (r) that represent the magnitude and direction of correlation with food patterns; e includes fish and chips, fish cakes, and fish fingers (crumbed, fried or deep fried); † Natural food sources of n-3 LCPUFA;†Natural sources of n-3 LCPUFA; †† Natural sources of n-3 ALA; * Products available in supermarket contained omega-3 (ALA or n-3 LCPUFA) 137

Chapter 5. Food pattern study

Table 3. Characteristics of Australian children within each food pattern group, by gender (age = 9-13 years, n = 1,110) Boys (n = 525) Snack fooda Soft drink Vegetables Pork and meat chops, steak Characteristics of children and mince (n = 121) (n = 156) (n = 135) (n = 113) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Age (y) 11 ± 1.4 11 (10, 12) 11.2 ± 1.3 11 (10, 12) 11 ± 1.4 11 (10, 12) 11 ± 1.4 11 (10, 12) BMI (kg/m2) 18.8 ± 3.2 17.9 (16.3, 20.3) 19.1 ± 3.6 18.4 (16.5, 21.3) 19.0 ± 3.5 18.4 (16.6, 20.5) 19.5 ± 3.7 18.6 (17.0, 21) Total n-3 LCPUFA intake (mg/d) 120 ± 282 40.8 (17.4, 84.9) 85.4 ± 137 35.1 (13.4, 100.7) 84.2 ± 183 30.8 (12.4, 72.1) 79.3 ± 179 30.9 (17.4, 79.8) Total n-3 LCPUFA intake 12.5 ± 29.5 4.7 (2, 10.2) 8.7 ± 12.2 4.4 (1.4, 10.9) 9.3 ± 19.4 3.5 (1.5, 8.6) 9.0 ± 22.8 3.7 (1.6, 8.2) [mg/d/energy (MJ)] Fish consumptionb (g/d) 13 ± 43 0 (0, 0) 3.7 ± 17.9 0 (0, 0) 14 ± 38 0 (0, 0) 16 ± 60 0 (0, 0) Take aways (meat, fish, cereal and 38 ± 99 0 (0, 0) 135 ± 188 59 (0, 239) 20 ± 68 0 (0, 0) 35 ± 72 0 (0, 40) vegetable mixed) consumption (g/d) Fish eater [n (%)] 27 (22.3) 30 (19.2) 25 (18.5) 17 (15) Girls (n = 585) Vegetables Take aways Tea, coffee, iced coffee drinks Canned meals and soup Characteristics of children (n = 130) (n = 167) (n = 129) (n = 159) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Mean ± SD Median (IQR) Age (y) 11 ± 1.5 11 (10, 12) 11 ± 1.5 11 (10, 13) 11 ± 1.4 11 (10, 12) 11 ± 1.3 11 (10, 12) BMI (kg/m2) 19.9 ± 3.7 19.1 (17.1, 22.5) 20.5 ± 4.8 20 (17.1, 22.5) 19.7 ± 3.8 18.8 (16.9, 21.9) 19.8 ± 4.1 18.8 (16.7, 21.7) Total n-3 LCPUFA intake (mg/d) 71.9 ± 161 31.5 (12.1, 61.8) 78.4 ± 175 29.4 (11.3, 72.5) 66.7 ± 114 30 (13.5, 72.9) 96.8 ± 198 40.8 (16.2, 83.3) Total n-3 LCPUFA intake 8.9 ± 17.5 4.1 (1.7, 8.5) 10 ± 20.2 3.7 (1.6, 8.3) 8.2 ± 13 4.1 (1.7, 8.9) 13.5 ± 30 5.4 (2.1, 10.4) [mg/d/energy (MJ)] Fish consumptionb (g/d) 32 ± 92 0 (0, 2.8) 5.4 ± 26 0 (0, 0) 12 ± 48 0 (0, 0) 4.3 ± 19 0 (0, 0) Take aways (meat, fish, cereal and 35 ± 99 0 (0, 0) 116 ± 212 0 (0, 153) 39 ± 100 0 (0, 2.7) 28 ± 76 0 (0, 0) vegetable mixed) consumption (g/d) Fish eater [n (%)] 19 (14.6) 35 (21) 19 (14.7) 32 (20.1) a Obtained from factor analysis with extraction method = principle component analysis, rotation method = varimax with Kaiser normalization and rotation converged in 15 iterations b Fish prepared from home (include canned fish and exclude fish bought from take-away outlets)

138

Chapter 5. Food pattern study

Association between food patterns and n-3 LCPUFA intake status

Discriminant analysis was conducted to predict how well the food patterns separated children according to n-3 LCPUFA intake status and fish eater status (Table 4). Differences in mean intake that approached statistical significance were observed in the ‘vegetable’ food pattern for boys who consumed n-3 LCPUFA levels similar to the AI (p = 0.067), and in the ‘take-away’ food pattern for girls who were fish eaters (p = 0.06). No food patterns were able to predict the ability of children to achieve the SDT for n-3 LCPUFA (Table 4).

Table 4. Relationship between food patternsa and n-3 LCPUFA intake status in Australian children (age = 9-13 years, n = 1,110) - Discriminant analysis (test t of equality of group means) Boys Food patterns Achieve p Achieve p Fish eater p SDTb AIb statusc Snack food 0.999 0.496 1.000 0.653 0.998 0.332 Soft drink 1.000 0.642 1.000 0.616 0.997 0.227 Vegetables 0.999 0.430 0.994 0.067 0.999 0.537 Pork and meat 0.999 0.428 0.997 0.211 0.998 0.335 chops, steak and mince Girls Food patterns Meet SDT p Meet AI p Fish eater p status Vegetables 1.000 0.834 0.999 0.415 1.000 0.678 Take aways 1.000 0.889 0.999 0.399 0.994 0.060 Tea, coffee, iced 0.998 0.292 0.996 0.145 0.997 0.200 coffee drink Canned meals 0.996 0.132 0.998 0.338 1.000 0.697 and soup a Obtained from factor analysis with extraction method = principle component analysis, rotation method = varimax with Kaiser normalization and rotation converged in 15 iterations b Wilks’ Lambda, the proportion of the total variance in the discriminant score not explained by differences among the groups

Discussion

In this nationally representative study of Australian children aged 9 to 13 y, food sources of n-3 LCPUFA together with foods that have the potential to be enriched in Australia were examined, in the context of dietary patterns in children. Factor analysis was used to generate food patterns in which correlations between consumption of various foods were captured, and subjects were grouped according to scores based on 139

Chapter 5. Food pattern study

diets with similar patterns of variation. Hence, major dietary habits can be characterized within a population [15].

The dietary patterns generated in this study support previous studies in Australian children, in which high food loadings have been reported for high fat and sugar intake [27, 28], vegetables [27], sugar sweetened beverages [29] and snack foods [30]. For example, a study of Western Australian adolescents demonstrated that their food patterns reflected a Western diet characterized by high consumption of take-away foods, soft drinks, confectionery, French fries, refined grains, full-fat dairy products and processed meats [28]. This pattern resulted in a high intake of saturated fat and monounsaturated fat, refined sugar and sodium, and low intake of fibre, folate and natural sugars, and was associated clustering of CVD risk factors, including raised total cholesterol levels [29]. A Western dietary pattern was also associated with poor behavior outcome scores, for example withdrawal, somatic complaints, anxiety/depression, delinquency and aggression, while a higher intake of leafy green vegetables and fresh fruit was correlated with better behavioral outcomes [31]. Another study reported that the intake of Victorian adolescents (12-13 y) did not conform to recommendations outlined by the Australian Guide to Healthy Eating (AGHE) [32], and that 22% reported eating fast foods every day, while over a third reported eating fruit rarely or never [33].

In the context of n-3 LCPUFA intake, the data in this study showed that girls in the ‘take-away’ food pattern and boys in the ‘soft drink’ food pattern more frequently consumed fish bought from outlets or restaurants, while boys and girls in the ‘vegetables’ food pattern, boys in the ‘pork and meat chops, steak and mince’ group and girls in the ‘take-away’ group more often consumed fish prepared from home (Table 2). About one fifth of boys and girls in these categories were fish consumers, on at least one of the days of the survey (Table 3). In contrast, children in the other food pattern groups rarely consumed fish, as demonstrated by a negative value of the correlation coefficients. Our data also demonstrated that consumption of ‘take aways’ in girls was likely associated with being a fish eater (p = 0.060). Fish is the major contributor to n-3 LCPUFA intake [34] and meat also contain n-3 LCPUFA [5]. Although overall fish intake was low in this study, the factor loadings of fish bought from take-away outlets in our study were approximately 4 times higher compared to fish prepared from home,

140

Chapter 5. Food pattern study

(Table 2) which could have nutritional implications. In a study among Australian families with young children aged 9-13 years, nearly half of the sample reported consuming take-away fish and chips at least once a month [35]. Although the frying method does not necessarily impact on the total content of EPA and DHA in the fish, the oil used for frying will change the lipid content of the fish due to oil absorption during frying. It has been reported that the use of an n-6 oil, such as sunflower or safflower, results in a decreased ratio of n-3 to n-6 in fried salmon [36,37] codfish, hake, sole [38] mackerel and sardines [37].

In this study, food consumption in the ‘vegetables’ food pattern group in both boys and girls tended to be healthier compared to the other food patterns, with a greater variety of plant-based foods. These included higher intakes of a range of common vegetables such as carrots, broccoli and green beans, and salad items such as tomato, cucumber and beetroot, together with sweet potato, potato, nuts and seeds. Vegetable consumption was associated with intake of chicken in girls, and sausages and soft drinks in boys, and also lower intakes of fish in both groups. Furthermore, of foods that have potential to be enriched with n-3 LCPUFA, eggs were consumed within the ‘vegetables’ pattern for both sexes (Table 2). Vegetables are recommended as part of healthy eating guidelines [32] and benefits of vegetable consumption are well documented, including prevention of food-related chronic diseases risk such as CVD [39]. In a study of American families with young children [40], vegetable consumption demonstrated a powerful role in increasing the enjoyment of family meals, not only as part of balanced meal but also as a flavor enhancer. It has been reported that children meeting the SDT for n-3 LCPUFA intake consume more vegetables (178 g) than children who do not meet the SDT (138 g) [7].

The data in this study confirms previous reports [10] that most Australian children do not consume fish (Table 2). A number of barriers to fish consumption have been identified in children, such as negative attitude towards both the smell and the accompaniments, and fear of finding bones [41], as well as the influence of other family members [42]. The level of confidence of the mother to serve fish for their family has been reported as a significant determinant of whether fish is included as a regular menu item in Australian families with young children [43]. We previously surveyed Australian families with young children, and demonstrated that cooking courses and

141

Chapter 5. Food pattern study

cooking books were scored as the most important nutrition education materials for improvement of the self-confidence of parents to serve fish for their family [35].

In countries with traditionally low fish consumption such as Australia, there are recommendations to consume more n-3 enriched products or to take fish oil supplements as alternative strategies to improve intake of n-3 LCPUFA [42]. The health benefits of consumption of products enriched with n-3 are well documented [43- 47] even with small dose enrichment [48]. Some land-based plants such as flaxseed can produce short chain n-3 oils, but are unable to produce the long chain n-3 oils such as DHA. Currently, the Commonwealth Scientific and Industrial Research Organization (CSIRO) are developing plants containing the n-3 LCPUFA typically found in fish oil using gene technology [49]. This could provide an alternative and more sustainable source of these oils for Australian to meet the dietary recommendation for n-3 LCPUFA.

A potential limitation of our study is that data were obtained from two 24-h dietary recalls. However, while this tool may not capture usual individual intake, it is appropriate for obtaining patterns of intake at a group level. Our results are also comparable with those of other studies that have reported low fish consumption among Australian children [50,51]. A further limitation is that factor analysis, while being recognized as a valid method to assess dietary patterns in a group, it is sample specific and therefore the results may not be generalizable to other populations.

Conclusion

Different food patterns between boys and girls were obtained in Australian children’s diet. Food patterns that include a high consumption of vegetables in boys and ‘take aways’ that include meat and fish in girls are likely to influence the total dietary n-3 LCPUFA intake. Nutrition education is needed to support the recommendation of two serving fish per week in order to meet dietary targets for n-3 LCPUFA intake, including information on appropriate cooking methods to maintain the health benefits of fish consumption associated with n-3 LCPUFA intake.

142

Chapter 5. Food pattern study

Acknowledgements

We would like to thank to the Australian Social Science Data Archive for permission using the 2007 Children’s survey dietary data. We would also like to thank the Directorate General of Higher Education Indonesia for sponsoring Setyaningrum Rahmawaty, a lecturer from the University of Muhammadiyah Surakarta Indonesia for her PhD at the University of Wollongong, New South Wales, Australia.

References

1 Russo GL. Dietary n-6 and n-3 polyunsaturated fatty acids: from biochemistry to clinical implication for cardiovascular prevention. Biochem Pharm 2009; 77: 937- 946.

2 Riediger ND, Othman RA, Suh M, Moghadasian MH. A systemic review of the roles of n-3 fatty acids in health and disease. J Am Diet Assoc 2009; 109: 668-679.

3 Burrows T, Collins CE, Garg ML. Omega-3 index, obesity and insulin resistance in children. Int J Pediatr Obes 2011; 6: e532-e539.

4 National Health Medical Research Council. Nutrient reference value for Australia and New Zealand including recommended dietary intakes. Australian Government Department of Health and Ageing. Commonwealth of Australia, 2006.

5 Howe P, Meyer B, Record S, Baghurst K. Dietary intake of long-chain ω-3 polyunsaturated fatty acids: contribution of meat sources. Nutrition 2006; 22: 47-53.

6 National Health Foundation of Australia. Position statement, fish, fish oils, n-3 polyunsaturated fatty acids and cardiovascular health, 2008. Available at: http://www.heartfoundation.org.au/sitecollectiondocuments/fish-position- statement.pdf : accessed on January, 15, 2011.

7 Meyer BJ, Kolanu N. Australian children are not consuming enough long-chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 2011; 27: 1136- 1140.

8 Sioen I, Huybrechts I, Verbeke W, Camp JV, Henauw SD. n-6 and n-3 PUFA intakes of pre-school children in Flanders, Belgium. Br J Nutr 2007; 98: 819-825.

9 Sichert-Hellert W, Wicher M, Kersting M. Age and time trends in fish consumption pattern of children and adolescents, and consequences for the intake of long-chain n- 3 polyunsaturated fatty acids. Eur J Clin Nutr 2009; 63: 1071-1075.

10 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ. Dietary intake and food sources of omega-3 long chain EPA, DPA and DHA of Australian children. Lipids 2013; 48: 869-877. 143

Chapter 5. Food pattern study

11 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ. Dietary modelling of n-3 LCPUFA: effect of replacement of bread, egg, milk and yoghurt with n-3 enriched for these foods on n-3 LCPUFA intake of Australian children. Nutrition 2013 – submitted.

12 Jacob DR, Gross MD, Tapsell LC. Food synergy: an operational concept for understanding nutrition. Am J Clin Nutr 2009; 89(suppl): 1543S–8S.

13 Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 2002; 13: 3-9.

14 Fung TT, Schulze M, Manson JE, Willett WC, Hu FB. Dietary patterns, meat intake, and the risk of type 2 diabetes in women. Arch Intern Med 2004; 164: 2235-2240.

15 Reedy J, Wirfält E, Flood A, Mitrou PN, Krebs-Smith SM, Kipnis V et al. Comparing 3 dietary pattern methods-cluster analysis, factor analysis, and index analysis-with colorectal cancer tisk. The NIH-AARP diet and health study. Am J Epidemol 2010; 171: 479-487.

16 Newby PK, Muller D, Hallfrisch J, Andres R, Tucker KL. Food patterns measured by factor analysis and antropometric changes in adults. Am J Clin Nutr 2004; 80: 504-513.

17 Venkaiah K, Brahmam GNV, Vijayaraghavan K. Application of factor analysis to identify dietary patterns and use of factor scores to study their relationship with nutritional status of adult rural populations. J Health Popul Nutr 2011; 4: 327-338.

18 Liorert S, McNaughton SA, Crawford D, Spence AC, Hesketh K, Campbell KJ. Parents’ dietary patterns are significantly correlated: findings from the Melbourne Infant Feeding Activity and Nutrition Trial (InFANT) Program. Br J Nutr 2011; 108: 518-526.

19 Northstone K. Dietary patterns: the importance of sex differences. Br J Nutr 2012; 108: 393-394.

20 Barrett-Connor E. Sex differences in coronary heart disease. Why are women so superior? The 1995 Ancel Keys Lecture. Circulation1997; 95: 252-264.

21 Stone NJ. Fish consumption, fish oil, lipids, and coronary heart disease. Circulation 1996; 94: 2337-2340.

22 Commonwealth of Australia. 2007 Australian National Children’s Nutrition and Physical Activity Survey-main findings. Commonwealth of Australia: 2008. Available at: http://www.health.gov.au/internet/main/publishing.nsf/content /66596E8FC68FD1A3CA2574D50027DB86/$File/childrens-nut-phys-survey.pdf: accessed on June, 23, 2010.

23 Australia Social Science Data Archive 2009. Available at: http://assdanesstar.anu. edu.au/webview/?object = http://assda-nesstar.anu.edu.au/ obj/fCatalog/Catalog28: accessed on January, 7, 2009.

144

Chapter 5. Food pattern study

24 Department of Health and Ageing (2007) User guide 2007 Australian national children’s nutrition and physical activity survey. Department of Health and Ageing. Available at: http://www.health.gov.au/internet/main/publishing.nsf/Content/phd- nutrition-childrens-survey-userguide: accessed on June, 23, 2010.

25 Allen P, Bennet K. SPSS for the health and behavioural sciences. Thomson; South Melbourne, Victoria, Australia 2008.

26 Venkaiah K, Brahmam GNV, Vijayaraghavan K. Application of factor analysis to identify dietary patterns and use of factor scores to study their relationship with nutritional status of adult rural populations. J Health Popul Nutr 2011; 29: 327-338.

27 McNaughton SA, Ball K, Mishra GD, Crawford DA. Dietary patterns of adolescents and risk of obesity and hypertension. J Nutr 2008; 138: 364-370.

28 Ambrosini GL, Huang RC, Mori TA, Hands BP, O’Sullivan TA, de Klerk NH et al. Dietary patterns and markers for the metabolic syndrome in Australian adolescents. Nutr Metab Cardiovasc Dis 2010; 20: 274-83.

29 Hafekost K, Mitrou F, Lawrence D, Zubrick SR . Sugar sweetened beverage consumption by Australian children: implications for public health strategy. BMC Public Health 2011; 11: art no 950.

30 Fayet F, Mortensen A, Baghurst K. Energy distribution patterns in Australia n its relationship to age, gender and body mass index among children and adults. Nutr Diet 2012; 69: 102-110.

31 Oddy WH, Robinson M, Ambrosini GL, O′Sullivan TA, de Klerk DH, Beilin LJ et al. The association between dietary patterns and mental health in early adolescence. Preventive Med 2009; 49: 39-44.

32 National Health and Medical Research Council (2013) Australian Dietary Guidelines. Canberra: National Health and Medical Research Council. Available at: http://www.nhmrc.gov.au/guidelines/publications/n55: accessed on March, 25, 2013.

33 Savige GS, Ball K, Worsley A, Crawford D. Food intake patterns among Australian adolescents. Asia Pac J Clin Nutr 2007; 17: 738-747.

34 Meyer BJ, Mann NJ, Lewis JL, Milligan GC, Sinclair AJ, Howe PR. Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids. Lipids 2003; 38: 391-8.

35 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ. Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Nutr Diet 2013, doi: 10.1111/1747-0080.12022.

36 Gladyshev MI, Sushchik NN, Gubanenko GA, Demirchieva SM, Kalachova GS. Effect of way of cooking on content of essential polyunsaturated fatty acids in muscle tissue of humpback salmon (Oncorhynchus gorbuscha). Food Chem 2006; 96: 446-451.

145

Chapter 5. Food pattern study

37 Candella M, Astiasarán I, Bello J. Deep-fat frying modifies high-fat fish lipid fraction. J Agri Food Chem 1998; 46: 2793-2796.

38 Candela M, Astiasarán I, Bello J. Effects of frying and warmholding on fatty acids and cholesterol of sole (Solea solea), codfish (Gadus morrhua) and hake (Merluccius merluccius). Food chem 1997; 58: 227-231.

39 Dauchet L, Amouyel P, Hercberg S, Dallongeville J. Fruit and Vegetable Consumption and Risk of Coronary Heart Disease: A Meta-Analysis of Cohort Studies. J Nutr 2006; 136: 2588-2593.

40 Wansink B, Shimizu M, Brumberg A. How vegetables make the meal: their hedonic and heroic impact on perceptions of the meal and of the preparer. Public Health Nutr 2012; 1-7.

41 Prell H, Berg C, Jonsson L. Why don’t adolescents eat fish? Factors influencing fish consumption in school. Scandinavian J Articles 2002; 46: 184-191.

42 McManus A, Burns SK, Howat PA, Cooper L, Fielder L. Factors influencing the consumption of seafood among young children in Perth: a qualitative study. BMC Public Health 2007; 7: 119, doi: 10.1186/1471-2458-7-119.

43 Mantzioris E, Cleland LG, Gibson RA, Neumann MA, Demasi M, James MJ. Biochemical effects of a diet containing foods enriched with n-3 fatty acids. Am J Clin Nutr 2000; 72: 42-48.

44 Murphy K, Meyer BJ, Mori TA, Burke V, Mansour J, Patch CS et al. Impact of foods enriched with omega-3 long chain polyunsaturated fatty acids on erythrocyte omega-3 levels and cardiovascular risk factors. Br J Nutr 2007; 97: 749-757.

45 Dangardt F, Osika W, Chen Y, Nilsson U, Gan LM, Gronowitz E et al. Omega-3 fatty acid supplementation improves vascular function and reduces inflammation in obese adolescents. Atherosclerosis 2010; 212: 580-515.

46 Lopez-Huertas E. Health effects of oleic acid and long chain omega-3 fatty acids (EPA and DHA) enriched milks. A review of intervention studies. Pharmacological Res 2010; 61: 200-207.

47 Romeo J, Wärnberg J, García-Mármol E, Rodríguez-Rodríguez M, Diaz LE, Gomez- Martínez S et al. Daily consumption of milk enriched with fish oil, oleic acid, minerals and vitamins reduces cell adhesion molecules in healthy children. Nutr Metab Cardiovasc Dis 2011; 21: 113-120.

48 Yep YL, Li D, Mann NJ, Bode O, Sinclair AJ. Bread enriched with microencapsulated tuna oil increases plasma docosahexaenoic acid and total omega-3 fatty acid in humans. Asia Pac J Clin Nutr 2002; 11: 285-291.

49 Commonwealth Scientific and Industrial Research Organization (CSIRO). Australian scientific collaboration set to break world’s reliance on fish for long chain omega-3.

146

Chapter 5. Food pattern study

Available at http://www.csiro.au/en/Organisation-Structure/Divisions/Plant- Industry/Omega3-Canola-Collaboration_PI.aspx: accessed on July, 12, 2012.

50 Clayton EH, Hanstock TL, Watson JF. Estimated intakes of meat and fish by children and adolescents in Australia and comparison with recommendations. Br J Nutr 2009; 101: 1731-1735.

51 Zhou SJ, Gibson RA, Gibson RS, Makrides M. Nutrient intakes and status of preschool children in Adelaide, South Australia. Med J Aus 2011; 196: 696-700.

147

Chapter 6. Validation study

Chapter 6

The majority of this section is the substantive content of the work submitted to Public Health Nutrition.

Statement

As the primary supervisor, I, Barbara J Meyer, declare that the greater part of the work in this paper is attributed to the candidate, Setyaningrum Rahmawaty.

In the manuscript, Setyaningrum contributed to study design and was primarily responsible for data collection, data analysis and data interpretation.

A/Prof Barbara J Meyer (Main Supervisor)

148

Chapter 6. Validation study

Development and validation of a food frequency questionnaire to assess n-3 LCPUFA intake in Australian children aged 9-13 years

(Submitted to Public Health Nutrition)

Setyaningrum Rahmawaty1,2*, Karen Charlton 2, Philippa Lyons-Wall3 and Barbara J Meyer1,2

1Metabolic Research Centre, University of Wollongong: 2School of Health Sciences, Faculty of Health Behaviour Sciences, University of Wollongong, Northfields Ave, Wollongong NSW 2522: 3now at School of Exercise and Health Sciences, Edith Cowan University, 270 Joondalup Drive, Joondalup, WA 6027

*Corresponding author: Email [email protected], Ph: +61 (0)2 4221 3459, Fax: +61 (0)2 4221 5945

Abstract

Objective: To develop a food frequency questionnaire (FFQ) aimed to assess dietary omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) intake of Australian children and to validate the FFQ against a 7-day food diary. Design: Cross-sectional and validation study. Setting: Two private primary schools in the Illawarra region of New South Wales. Subjects: Twenty two Australian children aged 9-13 years who were not on a special diet or receiving medical care that limited their food choice in the 3 months prior to recruitment. Results: A total of 131 items, classified according to 7 food group categories were included in the n-3 LCPUFA FFQ, as identified from published dietary surveys and a supermarket survey. Good correlations between the FFQ and 7d food diary were observed for eicosapentaenoic acid (EPA) (r = 0.691, 95 % CI = 0.51 – 0.83, p < 0.001), docosahexaenoic acid (DHA) (r = 0.684, 95 % CI = 0.45 – 0.84, p < 0.001) and total n-3 LCPUFA (r =0.687, 95 % CI = 0.48 – 0.85, p < 0.001). Bland-Altman plots showed an 149

Chapter 6. Validation study

acceptable limit of agreement between the FFQ and the average 7d food diary. However, the mean EPA, DHA and total n-3 LCPUFA intakes estimated from the FFQ were significantly higher than those from the average 7d food diary estimates (p < 0.001). Conclusion: A novel n-3 LCPUFA FFQ that has been developed to estimate dietary n-3 LCPUFA intakes in Australian children has been shown to have relative validity. The FFQ provides a useful contribution to dietary assessment methodology in this age group, however reproducibility remains to be demonstrated.

Keywords: omega-3 LCPUFA intake, FFQ, children, Australia, dietary assessment

Introduction

Omega-3 LCPUFA intakes including eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA) and docosahexaenoic acid (DHA) are essential for optimal health in children, particularly to support visual and cognitive development(1,2). The dietary sources of n-3 LCPUFA, however, are limited to a few commonly eaten foods, for example fish and seafood, meat, eggs and foods that have been enriched with n-3 LCPUFA(3).

The n-3 LCPUFA are insufficiently synthesized de-novo in the human body(4,5,6,7), therefore, to obtain enough of these fatty acids for optimal health, consumption of food sources containing preformed n-3 LCPUFA is warranted. Previous studies have demonstrated that fish consumption is significantly associated with the concentration of n-3 LCPUFA in human tissues such as serum(8,9,10,11), plasma(12,13) and erythrocyte(14), therefore it can be used as an indicator of n-3 LCPUFA status.

Since fish or seafood, the richest food sources of n-3 LCPUFA(3) is not regularly consumed by Australian children(15), the use of the 24-h recall, which is the method of choice for national nutrition surveys, to estimate habitual intake of this nutrient may lead to potential reporting bias. If the dietary report happens to be on a day that fish or seafood is not consumed, many fish or seafood consumers will report zero intakes on the 24-h recall. Therefore, supplementary information that covers frequency of fish and seafood

150

Chapter 6. Validation study

consumption over a longer reference period of time such as in the case of a food frequency questionnaire (FFQ) might improve estimates of n-3 LCPUFA intakes. Food frequency information could be a useful covariate in estimating the probability of consumption and amount of episodically consumed foods (e.g. foods that are not consumed nearly every day)(16). Therefore, it can enhance, not replace, dietary recalls data obtained in national survey, for example when estimating usual intake from a few day recalls using statistical modeling by taking into account day to day variation(16).

Currently, there is no FFQ specifically designed to assess n-3 LCPUFA intake in Australian children. A polyunsaturated fatty acid (PUFA) FFQ for use in Australian adults has been previously developed, and showed to be a valid and reproducible method to estimate n-3 LCPUFA intake, compared to biomarkers of n-3 LCPUFA (red blood cells and plasma fatty acids) and 3d weighed food records(17,18,19). The aims of this study were to develop and validate a FFQ to assess dietary n-3 LCPUFA intake in Australian children aged 9-13 years.

Material and methods

Subjects

Development and pilot testing of the FFQ

Australian children aged 9-13 years who were not on a special diet or receiving medical care that limited their food choice in the 3 months priors were recruited between May 2011 and December 2011. The children were recruited through poster advertisements that were distributed to parents or primary caregivers that participated in a survey to assess fish consumption behaviours(20) and enclosed letter sent to parents. Nine children were recruited this way and completed the pilot testing of the draft FFQ.

Validation of the FFQ

Primary school children aged 9-13 years were recruited between January and December 2012. Two schools were randomly selected from a total of 49 private schools in the Illawarra region, New South Wales, Australia (http://www.privateschoolsguide.com/ view- 151

Chapter 6. Validation study

users-list/wollongong-private-schools.html). The children were recruited through sealed letters sent to their parents through the school system. A total of 32 children were consented to participate in the study, however, only 22 children completed both the FFQ and 7d food diary.

Ethics approval was obtained from the Human Research Ethics Committee at the University of Wollongong. All subjects and their parents or primary caregivers provided written informed consent for participation in the study.

Development and a pilot study of the FFQ

Food items included in the development of the n-3 LCPUFA FFQ were selected from the following sources: a PUFA FFQ validated for use in Australian adults using red blood cell biomarkers(17,19), the 2007 Australian National Children’s Nutrition and Physical Activity Survey (Children’s Survey) nutrient file(21), a survey of fish consumption patterns of Australian children and their families (Fish Survey)(3) and an audit of foods enriched with n-3 LCPUFA that were available at major supermarkets in New South Wales in December 2009 (unpublished data) using the following steps:

1. Food items that contributed to total n-3 LCPUFA were identified from the PUFA FFQ(17), Children’s survey(21), Fish survey(20) and supermarket survey were manually aggregated and listed into the following seven food groups based on the richest food sources of the n-3 LCPUFA(15): (1) fish and seafood; (2) meat products and dishes; (3) egg products and dishes; (4) milk products and dishes; (5) cereal products and dishes; (6) spreads and margarines; (7) fish oil supplements. Categories for reporting of frequency of food consumption were based on a semi-quantitative FFQ, the Australian Child and Adolescent Eating Survey’ (ACAES)(22).

2. Commonly consumed portion sizes of the identified food sources of n-3 LCPUFA were identified using dietary data from the 1,110 children aged 9-13 years included in the nationally representative Children’s survey(21). A food model booklet was developed as an aid to assist children to quantify their portion sizes. Both the draft FFQ and food model booklet were piloted with input from academic nutrition professionals and

152

Chapter 6. Validation study

registered dieticians and pilot testing in five children; thereafter a final version was drafted and retested in an additional nine children. The outcomes of the second pilot test were used to remodify the final version of the FFQ. A supplementary questionnaire was completed by parents or care givers, and used as a cross-check for brands of foods and/or supplements consumed by the children.

Validation of the n-3 LCPUFA FFQ

The revised final version of the FFQ, called an n-3 LCPUFA FFQ, was used in the validation study, and a 7d food diary was used as a reference method.

7d Food diary

Each participant received a blank 7d food diary, with accompanying written instructions on how to record daily food consumption as well as any dietary supplements (e.g. fish oils). Participants were asked to record the names or brands and portion sizes of all food items and drinks consumed over a 7-day consecutive period. Any incomplete information was verified with the participants and their primary caregivers by the researcher.

Master database of n-3 LCPUFA

Dietary intake of total n-3 LCPUFA in Australian children from the 2007 Children’s Survey has been previously reported(15). From the same survey data, values of EPA, DPA and DHA were developed specifically to calculate fatty acids from the children’s diet (Judy Cunningham, 2009, personal communication); and called as a master database of n-3 LCPUFA to estimate n-3 LCPUFA intake.

Estimation of n-3 LCPUFA intake

Food items consumed by each child were translated into a daily equivalent frequency(23). Food quantities were calculated by multiplying the portion sizes (e.g. small, medium, large) with the daily equivalent frequency. The food frequency data were entered into FoodWorks dietary analysis package (Xyris Software Version 7.0.2959, 2012 Australia Pty Ltd) to identify food items (food ID) and the amount consumed by each child. Intake of EPA, DPA

153

Chapter 6. Validation study

and DHA were analyzed through merging the Food ID of the consumption data derived from the FoodWorks into n-3 LCPUFA master database. The n-3 LCPUFA content from supplements and foods that were not in the master database were added manually into the merged data. Total n-3 LCPUFA intake was calculated by summing the EPA, DPA and DHA values. Similar methods were applied to calculate EPA, DPA, DHA and total n-3 LCPUFA intakes from the 7d food diary, and total n-3 LCPUFA values were divided by 7 and expressed in milligrams (mg) per day.

Statistical analysis

Statistical analysis was carried out using Statistical Package for the Social Sciences (SPSS) software version 17.0, Chicago IL, USA. Intakes of EPA, DPA, DHA and total n-3 LCPUFA (mg/d) were presented as mean ± SD and median (inter quartile range, IQR). Dietary intakes of n-3 LCPUFA were not normally distributed therefore the data were transformed (Log10 ) for these analysis. The paired sample t-test was used to assess differences in n-3 LCPUFA intakes between the FFQ and the 7d food diary (mg/d). The Pearson’s correlation coefficient was used to assess the linear proximity relation(24) between FFQ and the 7d food diary. The correlation coefficients (r) are interpreted as follows: r ≤ 0.35 indicate weak correlation, r = 0.36 to 0.67 indicate moderate correlation, r = 0.68 to 1 indicate good correlation with r ≥ 0.9 indicate very good correlation(25,26). The bootstrap sampling technique was used to obtain confidence intervals (CI) which are computed as the 5th and 95th percentiles(27). The total of 1000 bootstrap samples of equal size (n = 22) was obtained from 22 subjects by random sampling with replacement. The MathWorks (MATLAB) software version R2011b (7.13.0.564), US was used to construct the bootstrapping sampling.

The Bland and Altman plot,(28,29) expressed as the mean difference and two standard deviations between the n-3 LCPUFA intakes obtained by the FFQ and the 7d food diary, was used to quantify limits of agreement (LA) between the two methods. If 95 % of the differences are within ± 1.96 standard deviations of the mean of the differences, this indicates good agreement between both measurements (28,29). The LA is defined such that we expect that, in the long run, 95 % of future differences between the measurements made

154

Chapter 6. Validation study

on the same subject will lie within these limits(30). The log-transformed LA values of the Bland Altman were transformed back by exponentiation which correspond to the ratio of one method’s measurements to the other(30) to provide an easy and reliable practical interpretation. Coefficient reproducibility, how well the FFQ can predict n-3 LCPUFA intakes were calculated. P values < 0.05 were considered significant.

Results

Mean age of total children participated in this study was 10 (SD ± 0.93) years, with 55 % of the sample comprising girls (n = 17). An n-3 LCPUFA FFQ developed in this study was consisted of 7-item food group consisted of a total of 131 items. The food-items included in the final n-3 LCPUFA FFQ are shown in Table 1.

Mean dietary n-3 LCPUFA estimated using the FFQ and 7d food diary are shown in Table 2. Mean ± SD (mg/d) dietary EPA, DHA and n-3 LCPUFA from FFQ were significantly higher, approximately 2, 2 and 1.7-fold, respectively, than estimates using a 7d food diary, but no difference between methods was found for DPA (Table 2).

Good significant correlations between the FFQ and 7d food diary were found for EPA, DHA and total n-3 LCPUFA (Table 3). Bland Altman plots showed an acceptable 95 % LA in logged measurement of the two methods. The mean difference between methods falls within the 95% CI for mean of difference (Table 4). There was no systematic variation between the two methods for EPA, DPA, DHA and total n-3 LCPUFA (Figure 1 to 4). Exponentiating the LA values gave the mean difference (and 95 % LA) for the ratio of the FFQ compared to the 7d menu diary were of 1.31 (ei. 31 % higher) (0.74 (26 % lower) to 2.30 (130 % higher)) for EPA; 1.11 (0.61 to 2.0) for DPA; 1.68 (0.78 to 3.61) for DHA and; 1.35 (0.81 to 2.27) for total n-3 LCPUFA.

155

Chapter 6. Validation study

Table 1. Food items assessed in the n-3 LCPUFA FFQ

Group 1: Fish & seafood Group 3: Egg products & Dishes Omega-3 enriched bread Canned fish and seafood - Eco eggs with Omega-3 - Uncle Toby's Plus Omega-3 Lift - Tuna - Pace Farm Omega-3 Free Range - Tip Top UP Sandwich bread with - Salmon Body eggs Omega-3 DHA - Herring - Farm Pride Free Range Omega-3 - Wonder White+Omega-3 DHA - Sardines - Veggs for Families eggs bread - Mackerel - Normal chicken eggs - COLES High Top bread with n-3 - Anchovies (including free range) - Tip Top UP Muffin with Omega-3 - Not sure of type - Other Omega-3 enriched eggs DHA Take-away fish products (please specify brand) - Diego Wrap Omega-3 - Take-away (e.g. fish & chips) - Duck eggs - McCain's Healthy Choice Pizza - Sushi - I don’t know the type with Omega-3 DHA - Fish burger - Other cereals/breads/muffins/wraps - Fish finger Group 4: Milk Products & Dishes added withomega-3 (please specify - Prawn cracker - Pura Kids Omega-DHA the brand) - Fish cake - Farmer's Best source of Omega-3 - I don’t know the type - Seafood stick - Other milk added with omega-3, Cereal-based products and dishes - Calamari (please specify the brand) - Savoury biscuits-crackers - Other (please describe) - Whole milk - Sweet biscuits, cookies Fresh or frozen fish - Fat reduced/lite milk - Cakes, pies, pancakes, custard, - Fish finger - Skim milk doughnuts, buns, pikelets, scones, - Tuna - Soy milk lamington, waffles, tarts - Barramundi - None of the above (please - Pudding - Blackfish specify) - Bream Milk products Group 6: Spread or margarine - John Dory - Flavored milk, e.g. hot chocolate, - Meadow Lea Hi-Omega canola - Snapper milkshake, smoothie, Milo drink spread - Swordfish - Plain milk - glass or with cereal - John West Tuna Fish Spread - Not sure of type - Yogurt, frozen yogurt - Seachange Omega-3 spread Lite Fresh or frozen seafood - Cheese-including on sandwiches, - Other Omega-3 enriched spread - Prawn toast or biscuits (please specify brand) - Squid/Calamari - Ice cream - Butter - Scallop - Yakult - Olive oil spread or blend - Oyster - Cream or sour cream - Dairy spread or blend - Lobster - Canola spread or blend - Octopus Group 5: Cereal Products & Dishes - Polyunsaturated spread or blend - Morton Bay Bugs Cereal-based products and dishes (e.g. Meadow Lea) - Crab/Seafood stick - Sandwich, roll - I don't know the type - Fish roe/fish egg - Cereal, e.g. breakfast cereal, corn - Yabbie flakes Group 7: Supplement - Not sure of type - Bread, e.g. bread rolls, pita, toast, Fish oil naan - Blackmores Kids fruity fishies Group 2: Meat Products & Dishes - Pasta, e.g. lasagna, spaghetti, - Children’s Omega-3 (chewy Meat bolognaise, ravioli, cannelloni capsules) - Sausages, e.g. salami, kabana, - Pizza - Haliborange Kids Omega-3 devon, saveloy, cabanossi, - Rice, e.g. fried rice, risotto - I.Q. Essentials Children Chewable pepperoni, etc. - Noodles Omega-3 Capsules - Pork, bacon, ham - English muffin, bagel, crumpet - Kids smart Omega3 Fish Oil - Beef, lamb, mutton - Wrap - Omeguard Kids’ - Turkey, chicken - Kebab - Swisse Children's Smart Fish Oil - Not sure the type - Oat, e.g. oatmeal, porridge 60 Chewable Capsules - Thompson’s Goldfish Junior Omega 3 Capsules - Triple Omega-3 Gummie Fish - Other fish oil capsules/omega-3 (please specify the brand) - I don’t know the type

156

Chapter 6. Validation study

Table 2. Comparison of EPA, DPA, DHA and total n-3 LCPUFA obtained from FFQ and the 7d food diary†

n-3 LCPUFA FFQ 7d food diary p value Mean SD Median Mean SD Median (IQR) (IQR) EPA 67.28 41.19 56.52 45.60 45.89 29.44 < 0.001 (mg/d) (31.35-89.80) (14.43-68.70) DPA 40.79 17.71 40.58 33.80 20.12 28.99 0.127 (mg/d) (28.66-56.74) (18.59-46.08) DHA 134.53 109.83 78.30 65.22 87.58 36.40 < 0.001 (mg/d) (63.76-195.06) (11.48-71.22) Total n-3 242.60 153.65 182.82 144.62 143.57 98.33 < 0.001 LCUFA (125.79-337.45) (48.89-184.20) (mg/d)

EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; Total n-3 LCPUFA, sum EPA+DPA+DHA; FFQ, food frequency questionnaire; IQR, inter quartile range; †Average of 7d; * Paired t- test after log transformation (Log10)

Table 3. Validity correlation between FFQ and the average 7d food diary

Pearson’s correlation 95 % CIa p value coefficient EPA 0.691c 0.51 – 0.83 < 0.001 DPA 0.204 -0.25 – 0.62 0.362 DHA 0.684 0.45 – 0.84 < 0.001 Total n-3 LCPUFA 0.687 0.48 – 0.85 < 0.001 aCI, confidence interval obtained using bootstrapping with 1000 bBias factor is the proportional bias in a univariate regression with the value intakes obtained from the FFQ were considered as the independent variable c Values after log transformation (Log10)

157

Chapter 6. Validation study

Table 4. Log transformed limit of agreementa between FFQ and the average 7d food diary

Limit of Mean of Standard 95 % CI for Coefficient agreement log difference error of mean mean of reproducibility transformed of difference difference (%) EPA -0.29 – 0.83 0.27 0.013 0.24 – 0.29 68.98 DPA -0.49 – 0.69 0.10 0.013 0.07 – 0.13 67.32 DHA -0.25 – 1.28 0.52 0.018 0.48 – 0.55 81.00 Total n-3 -0.21 – 0.82 0.30 0.012 0.28 – 0.33 62.72 LCPUFA aBland-Altman test CI, confidence interval; EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; Total n-3 LCPUFA, sum EPA+DPA+DHA

Fig. 1. Bland-Altman plot: comparison of the agreement of dietary EPA intakes determined using food frequency questionnaire and 7d food diary, after log transformation

158

Chapter 6. Validation study

Fig. 2. Bland-Altman plot: comparison of the agreement of dietary DPA intakesdetermined using food frequency questionnaire and 7d food diary, after log transformation

Fig. 3. Bland-Altman plot: comparison of the aggreement of dietary DHA intakesdetermined using food frequency questionnaire and 7d food diary, after log transformation

159

Chapter 6. Validation study

Fig. 4. Bland-Altman plot: comparison of the aggreement of dietary total n-3 LCPUFA intakes determined using food frequency questionnaire and 7d food diary, after log transformation

Discussion

We have developed a novel FFQ that is specifically designed to assess n-3 LCPUFA intake in Australian children aged 9-13 years and that has been demonstrated to have good relative validity. Any FFQ needs to be sensitive to cultural and dietary practices and is therefore population-specific. In addition, because of the difficulties that children experience in accurately recalling past dietary intakes, validation of a dietary instruments for use in the relevant age group is also required(31).

An accurate dietary instrument to assess n-3 LCPUFA intake in children has a number of uses. These include a reliable method to estimate habitual intake of n-3 LCPUFA intakes which, in turn, can be used to predict suboptimal n-3 LCPUFA intake- related diseases risks, monitoring trends in intakes of children, and informing the development of nutrition intervention programs and national policy. Therefore, the selection of an appropriate dietary instrument which can capture the consumption of food sources containing the n-3 LCPUFA is needed.

160

Chapter 6. Validation study

Ideally, biomarkers may be used as reference methods to validate a FFQ that is designed to assess n-3 LCPUFA intake, as has been reported in validation studies in adults(17,18 ,32-35). However, drawing of blood samples is invasive for children. An alternative option is using multiple days of food recalls such as in the 7d food diary(36-38). Multiple days of observation are needed to estimate habitual intake in order to improve the reliability of the measurement due to day today variation in intake patterns(39). It has been argued that 7d food records are an acceptable reference method for validation studies in children(40), because of the capability of the method to increase the precision of the intake data, as compared to one and 3 different interview days(40). Beside, it is lower respondents’ burden than weighed food diaries(41).

Mean intake of total n-3 LCPUFA that was calculated using this novel FFQ was approximately one and a half-fold higher than estimates obtained from a 7d food diary, while DHA intake was twice as high using the FFQ data. This discrepancy is explained by the longer reference period of reporting (i.e. one month) included in the FFQ, which resulted in fish consumption being captured by the FFQ in some children that had not consumed this food group over the past seven days. A reference period longer than one week may be required to obtain an estimate of habitual n-3 LCPUFA intake, especially from fish or seafood sources. Most children in our study reported consuming take-away fish and chips and home-made fish products on average between one to three times per month. This is similar to dietary patterns reported in our previous study of Australian families with young children in which nearly half of those who consumed fish reported purchasing take- away fish and chips at least once a month(20). National dietary surveys that commonly use two 24-h recalls may underestimate n-3 LCPUFA intake in children, since fish and seafood are not regularly consumed in this age group (15,20,42,43). For example, in the most recent nationally representative nutrition survey of Australian children (2007 Children’s survey)(44), average intakes of EPA (23.5 mg/d), DHA (40.9 mg/d) and total n-3 LCPUFA (85.3 mg/d) were approximately three-fold lower than those reported in our study, while DPA intakes, provided mostly from meat, were half the values(44).

In general, it has been reported that FFQs tend to overestimate actual intake, compared to 7d food records(45) as well as 7d weighed records(46). A study of Belgian 161

Chapter 6. Validation study

school-aged children (12 - 15 years) showed that a FFQ overestimated mean consumption for most foods (11 of 14 food items) including fruit, vegetables, bread, milk, chips, sweets, and soft drink, as compared to records from a 7d food diary. In that relative validity study, Spearman correlation coefficients ranged between 0.10 and 0.65 for various nutrients(45). A validation study among Australian rural children aged 10 to 12 years has reported a weak correlation for frequency of fish consumption between a short FFQ (mixed FFQ) and three repeated 24-h recalls (Kendal τ ≥ 0.1)(47).

Dietary intake measurements are mostly dependent on individual cognitive aspects, and the intakes values obtained must be considered as values of a latent variable(31). Capabilities in memorizing foods that have been consumed, as well as interpretation of portion size, differ between individuals, and error are probable, especially in children(41).

Children’s self reporting of dietary intakes are complex tasks, involving individual cognitive aspects in relation to perception, conception and memory(48). Children’s ability to accurately recall an amount of food eaten can improved through the use of portion-size aids, such as food drawings, models and photographs(49). Hence, designing a food frequency questionnaire for children requires numerous strategies to improve their memory and portion size estimation(50). It has been reported that children’s estimates of portion size using age-appropriate food photographs were significantly more accurate (an underestimation of 1 % on average) than estimates using photographs designed for use with adults (which overestimated by 45 % on average)(51). In our study, a food model booklet was developed to help children to recall the foods they had consumed. A supplementary questionnaire was completed by parents or care givers, and used as a cross-check for brands of foods and/or supplements consumed by the children.

This study has a number of limitations. Only one measurement per participant per method was made in this study, therefore, reliability or interclass correlation coefficient(30,52) cannot be compared(30). A limitation of relative validation is that both instruments may have the same source of random error, and therefore both may provide an incorrect assessment of a nutrient intake, but still be shown to have close agreement(31,53). For this reason, the method of triads (provide reference) which includes an appropriate

162

Chapter 6. Validation study

biomarker is the preferred analysis for validation of a FFQ to assess n-3 LCPUFA intakes(19,32-35,54). Few such studies have been conducted in children because of difficulty in obtaining biomarker variables in this age group. A USA study validated a FFQ completed by parents of children aged 1-11 years against children’s erythrocyte membrane concentrations of n-3 LCPUFA(54). Both total omega-3 fatty acids and marine PUFAs, expressed as percentage of total dietary fat, were associated with percent of omega-3 and marine PUFAs in the erythrocyte membranes(54).

In conclusion, the new FFQ developed in this study demonstrates relative validity in assessment of n-3 LCPUFA intakes in Australian children. Further research is needed to optimize the FFQ in order to limit overestimation, as well as to validate the FFQ against biomarkers of n-3 LCPUFA status and to assess reproducibility of the method at different time points.

Acknowledgements

We would like to thank to the participants and the private schools which participated in this study. Professor Clare E Collins – Clare, School of Health Sciences, Faculty of Health, University of Newcastle Australia is thanked for permission in adopting the frequency items of the ACAES. We also thank to the Directorate General Higher Education Indonesia for sponsoring Setyaningrum Rahmawaty, a lecturer from the University of Muhammadiyah Surakarta Indonesia for her PhD at the University of Wollongong, New South Wales, Australia. No competing interests were identified.

References

1 Eilander A, Hundscheid DC, Osendarp SJ et al. (2007) Effects of n-3 long chain polyunsaturated fatty acid supplementation on visual and cognitive development throughout childhood: A review of human studies. Prostaglandins Leukot Essent Fatty Acids 76, 189-203.

2 Jacques C, Levy E, Muckle G et al. (2011) Long-term effects of prenatal omega-3 fatty acid intake on visual function in school-age children. J Pediatr 158, 73–80.

3 Meyer BJ, Mann NJ, Lewis JL et al. (2003) Dietary intakes and food sources of omega- 6 and omega-3 polyunsaturated fatty acids. Lipids 38, 391-398. 163

Chapter 6. Validation study

4 Emken EA, Adlof RO, Gulley RM (1994) Dietary linoleic acid influences desaturation and acylation of deuterium-labeled linoleic and linolenic acids in young adult males. Biochim Biophys Acta 1213, 277- 288.

5 Pawlosky RJ, Hibbeln, Novotny JA et al. (2001) Physiological compartmental analysis of α-linolenic acid metabolism in adult humans. J Lipid Res 42, 1257-1265.

6 Burdge GC, Finnegan YE, Minihane AM et al. (2003) Effect of altered dietary n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13 C]α-linolenic acid to longer-chain fatty acids and partitioning towards β-oxidation in older men. Br J Nutr 90, 311-321.

7 Vlaardingerbroek H, Hornstra G, de Koning TJ et al. (2006) Essential polyunsaturated fatty acids in plasma and erythrocytes of children with inborn errors of amino acid metabolism. Mol Genet Metab 88, 159-165.

8 Amiano P, Dorronsoro M, M de Renobales et al. (2001) Very-long-chain ω-3 fatty acids as markers for habitual fish intake in a population consuming mainly lean fish: the EPIC cohort of Gipuzkoa. Eur J Clin Nutr 55: 827-832.

9 Oddy WH, Sherriff JL, Kendall GE et al. (2004) Patterns of fish consumption and level of serum phospholipid very-long-chain omega-3 fatty acids in children with and without asthma, living in Perth, Western Australia. Nutr Diet 61, 30-37.

10 Yep YL, Li D, Mann NJ et al. (2002) Bread enriched with microencapsulated tuna oil increases plasma docosahexaenoic acid and total omega-3 fatty acids in humans. Asia Pac J Clin Nutr 11, 285-291.

11 Gillingham LG, Caston L, Leeson S et al. (2005) The effects of consuming docosahexaenoic acid (DHA)-enriched eggs on serum lipids and fatty acid composition in statin-treated hypercholesterolemic male patients. Food Res International 38, 1117- 1123.

12 Baró L, Fonallá J, Peña JL et al. (2003) n-3 fatty acids plus oleic acid supplemented milk reduces total and LDL cholesterol, homocysteine and levels of endothelial adhesion molecules in healthy humans. Clin Nutr 22, 175-182.

13 Garg ML, Blake RJ, Clayton E et al. (2007) Consumption of an n-3 polyunsaturated fatty acid-enriched dip modulates plasma lipid profile in subjects with diabetes type II. Eur J Clin Nutr 61, 1312-1317.

14 Murphy K, Meyer BJ, Mori TA et al. (2007) Impact of foods enriched with omega-3 long chain polyunsaturated fatty acids on erythrocyte omega-3 levels and cardiovascular risk factors. Br J Nutr 97, 749-757.

15 Meyer BJ, Kolanu N (2011) Australian children are not consuming enough long-chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 27, 1136-1140. 164

Chapter 6. Validation study

16 Subar AF, Dodd KW, Guenther PM et al. (2006) The food propensity questionnaire: concept, development, and validation for use as a covariate in a model to estimate usual intake. J Am Diet Assoc 106, 1556-1563.

17 O’Sullivan BL, Williams BG, Meyer BJ (2006). Biomarker Validation of a Long-Chain Omega-3 Polyunsaturated Fatty Acid Food Frequency Questionnaire. Lipids 41, 845- 850.

18 O’Sullivan BL, Brown J, Williams PG et al. (2008) Dietary validation of a new food frequency questionnaire that estimates long-chain omega-3 polyunsaturated fatty acids. Br J Nutr 99, 660-666.

19 Swierk M, Williams PA, Wilcox J et al. (2011) Validation of an Australian electronic food frequency questionnaire to measure polyunsaturated fatty acid intake. Nutrition 27, 641-646.

20 Rahmawaty S, Charlton K, Lyons-Wall P et al. (2013) Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Nutr Diet, DOI: 10.1111/1747-0080.12022

21 Australia Social Science Data Archive (2009) Available at: http://assdanesstar.anu.edu.au /webview/?object = http://assda-nesstar.anu.edu.au/obj/ fCatalog/Catalog28. Accessed January, 7, 2009

22 Watson JF, Collins CE, Sibbritt DW et al. (2009) Reproducibility and comparative validity of a food frequency questionnaire for Australian children and adolescents. Int J Behav Nutr Phys Act 6, 62, doi:10.1186/1479-5868-6-62.

23 Cancer Council Victoria (2012) Dietary questionnaire for epidemiological studies version 2 (DQES v2), User information guide. Available at: http://www.cancervic. org.au\dges: accessed on September 2012.

24 Taylor BN, Kuyatt E (1994) Guidelines for evaluating and expressing the uncertainty of NIST measurement results. National Institute for Standards and Technology. Available at: http://physics.nist.gov/Pubs/guidelines/TN1297/tn1297s.pdf: accessed on April 2013.

25 Weber JC,Lamb DR (1970) Statistics and Research in Physical Education. St. Louis: CV Mosby Co, pp. 59-64.

26 Mason RO, Lind DA, Marchal WG (1983) Statistics: An Introduction. New York: Harcourt Brace Jovanovich, Inc, pp. 368-383.

27 Efron B, Tibshirani RJ (1993) Introduction to the Bootstrap. Chapman & Hall/CRC: New York.

165

Chapter 6. Validation study

28 Bland JM, Altman DG (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1, 307-310.

29 Bland JM, Altman DG (2003) Applying the right statistics: analyses of measurement studies. Ultrasound Obstet Gynecol 22, 85-93.

30 Bartlett JW, Frost C (2008) Reliability, repeatability and reproducibility: analysis of measurement errors in continuous variables. Utrasound Obstet Gynecol 31, 466-475. 31 Kaaks R, Ferrari P (2006) Dietary intake assessments in epidemiology: can we know what we are measuring? Ann Epidemiol 16, 377-380.

32 Mikkelsen T, Osler M, Olsen S (2006) Validity of protein, retinol, folic acid and n-3 fatty acid intakes estimated from the food-frequency questionnaire used in the Danish National Birth Cohort. Public Health Nutr 9, 771-778.

33 Mina K, Fritschi L, Knuiman M (2007) A valid semi quantitative food frequency questionnaire to measure fish consumption. Eur J Clin Nutr 61, 1023-1031.

34 McNaughton S, Hughes M, Marks G (2007) Validation of a FFQ to estimate the intake of PUFA using plasma phospholipid fatty acids and weighed foods records. Br J Nutr 97, 561-568.

35 Zhang B, Wang P, Chen C et al. (2010) Validation of an FFQ to estimate the intake of fatty acids using erythrocyte membrane fatty acids and multiple 3d dietary records. Public Health Nutr 13, 1546-1552.

36 Bandini LG, Cyr H, Must A et al. Validity of reported energy in preadolescent girls. Am J Clin Nutr 65, 1138S-1141S.

37 Vereecken CA, Maes L (2003) A Belgian study on the reliability and relative validity of the Health Behaviour in School-Aged Children food-frequency questionnaire. Public Health Nutr 6, 581-588.

38 Bloom L, Lundmark K, Dahlquist G, Persson LA (1989) Estimating children‘s eating habits:Validity of a questionnaire measuring food frequency compared to a 7-day record. Acta Prediatr Scand 7, 858-864.

39 Beaton GH, Milner J, Corey P et al. (1979). Sources of variance in 24-hour dietary recall data: implications for nutrition study design and interpretation. Am J Clin Nutr 179, 2546-2559.

40 Persson LÅ, Carlgren G (1984) Measuring children’s diet: Evaluation of dietary assessment techniques in infancy and childhood. Int J Epidemol 13, 506-517.

41 Wrieden W, Peace H, Armstrong J et al. (2003) A short review of dietary assessment methods used in National and Scottish Research Studies. Available at:

166

Chapter 6. Validation study

http://www.food.gov.uk/multimedia/pdfs/scotdietassessmethods.pdf.: accessed on June 2013.

42 English R, Cashel KM, Lewis JL et al. (1988) National Dietary Survey of Schoolchildren aged 10–15 Years. Report no. 1. Foods Consumed. Canberra: Australian Government Publishing Service.

43 McLennan W, Podger A (1999) National Nutrition Survey – Foods Eaten, Australia 1997 (Cat. No. 4802.0). Canberra: Australian Bureau of Statistics, Commonwealth Department of Health and Aged Care. Available at: http://www.ausstats.abs.gov.au/ ausstats/free.nsf/0/236465EA4E9B3D2BCA25722500049629/$File/48020_1995.pdf: accessed on January 2013.

44 Rahmawaty S, Charlton K, Lyons-Wall P et al. (2013) Dietary intake and food sources of omega-3 long chain EPA, DPA and DHA of Australian children. Lipids, in press.

45 Burrows T, Berthon B, Garg ML et al. (2012) A comparative validation of a child food frequency questionnaire using red blood cell membrane fatty acids. Eur J Clin Nutr 66, 825-829.

46 Pampaloni B, Bartolini E, Barbieri M et al. (2013) Validation of a Food-Frequency Questionnaire for the Assessment of Calcium Intake in Schoolchildren Aged 9–10 Years. Springer Science+Business Media New York, doi: 10.1007/s00223-013-9721-y.

47 Gwynn JD, Flood VM, D’Este CA et al. (2010) The reliability and validity of a short FFQ among Australian Aboriginal and Torres Strait Islander and non-Indigenous rural children. Public Health Nutr, 1-14, doi:10.1017/S1368980010001928.

48 Nelson M, Atkinson M, Darbyshire S (1994) Food photography I: The perception of food portion size from photographs. Br J Nutr 72, 649-663.

49 Subar AF, Crafts J, Zimmerman TP, Wilson M et al. (2010) Assessment of the accuracy of portion size reports using computer-based food photographs aids in the development of an automated self-administered 24-hour recall. J Am Diet Assoc 110, 55-64.

50 Baranowski T, Domel SB (1994) A cognitive model of children’s reporting of food intake. Am J Clin Nutr 59 (suppl), 212S-217S.

51 Foster E, Matthews JN, Nelson M et al. (2006) Accuracy of estimates of food portion size using food photographs - the importance of using age-appropriate tools. Public Health Nutr 9, 509-14.

52 Davis JA, Rozenek R, Decicco DM et al. (2007) Comparison of three methods for detection of the lactate threshold. Clin Physiol Funct Imaging 27, 381-384.

167

Chapter 6. Validation study

53 Ocke MC, Kaaks RJ (1997) Biochemical markers as additional measurements in dietary validity studies: application of the method of triads with examples from the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr 65, 1240S–1245S.

54 Orton HD, Szabo NJ, Clare-Salzler M et al. (2008) Comparison between omega-3 and omega-6 polyunsaturated fatty acid intakes as assessed by a food frequency questionnaire and erythrocyte membrane fatty acid composition in young children. Eur J Clin Nutr 62, 733-738.

168

Chapter 7

S u m m a r y

169

Chapter 7. Summary

7.1 Conclusions

7.1.1 Contextualising the thesis

Monitoring and surveillance of food and nutrition on a population level is fundamental to informing health policy decision and priorities of intervention programs. An understanding of food intake patterns and nutrient status of various sectors of the population, as in children, will inform the development and implementation of effective nutrition interventions that may result in changes in food behaviour to lead to healthier diets. Such strategies may be required to increase intakes of n-3 LCPUFA at the population level. The data produced through a systematic and ongoing process of dietary monitoring allows analysis and interpretation of indicators and causal determinants of food eating habit as well as environmental factors that may impact on access to sources of the nutrients and foods of interest. Such data would be useful to identify whether nutrient intakes of groups are in adequate or in the case of some nutrients, excessive. The use of nutritional supplements could be identified, as well as barriers and mediators to healthy eating practices, changes of food eating patterns associated to nutrition-related disease risks, updating culturally and socially appropriate food and dietary guides. Food monitoring data also could be used to enforce food regulation and guide food industry research and development and innovation of new products1. Therefore, periodic nutrition monitoring is needed not only to provide up-to- date information about the nutritional status of populations, but also to assess the impact of the efforts or recommendations to improve nutrition intake status of the community.

Health benefits of n-3LCPUFA in children, especially for DHA and EPA, are well documented with regard to brain and cognitive development, as well as the prevention of CVD risk factors in later life (Chapter 1). Therefore, monitoring levels of intake of n-3 LCPUFA is necessary, not only to measure whether intakes of these fats meet recommended requirements, but also to assess whether targets levels to reduce the development of chronic disease risk are being met. To date, there is little information available on intakes of the individual n-3 LCPUFAs, EPA, DPA and DHA, in the diets of Australian children. Moreover, the prevalence of nutrition-related chronic disease risks such as overweight and obesity which are known as an independent risk factor for later CVD risk2,3 tend to increase with increasing age in Australian children4,5.

170

Chapter 7. Summary

In order to assess and evaluate the intakes of EPA, DPA and DHA in Australian children, secondary analysis was undertaken of the latest nationally representative dietary survey, the Australian National Children’s Nutrition and Physical Activity (Children’s Survey)6. Food consumption data was obtained through two repeated 24-h dietary recalls. Findings from this analysis were used as the foundation on which to develop the sequence of subsequent studies in this thesis.

7.1.2 Core of thesis findings

Omega-3 LCPUFA intakes in Australian children were generally low, compared to the recommended intake for optimal health (Suggested Dietary Target)7,8. The low level of EPA and DHA intakes contributed significantly to the low intake of total n-3 LCPUFA, and was explained by low fish consumption. Although fish was the main food source for total n-3 LCPUFA (particularly for EPA and DHA), and children who consumed fish approximately 70 % of them meet the SDT, but only 20 % of children consumed fish regularly (Chapter 2)9. Furthermore, only less than 7 % of children consumed foods enriched with n-3 LCPUFA and fish oil supplements on either day of the surveys.

Given that consuming 2-3 serves of fish/week (preferably oily fish), together with foods enriched with n-3 LCPUFA, are recommended for optimal health10 and wellbeing11, a stepwise approach using intervention mapping12 was applied to construct potential strategies that can be used to improve children’s intake of these foods. Previous studies have shown that parental lifestyle influences children’s preference for healthy food. Parents or primary caregivers are considered to be the gatekeeper of family food supply; therefore, an approach directed to parents instead of children is recommended, especially when designing interventions aimed to improve dietary patterns. Since limited information was available about parental behaviour in the purchase and preparation of fish and/or seafood, and foods enriched with n-3 for their family, a survey was conducted to fill this gap in the literature. Chapter 3 describes the factors that encourage or prevent parents from purchasing and serving meals containing fish and/or seafood, as well as n-3 enriched food products. Most parents had good awareness of the health benefits of fish; however, only around 20 % reported serving fish regularly for their family. The smell, presence of bones, and possibly pollutants in fish, as well as inadequate cooking skills were considered as the main barriers. In

171

Chapter 7. Summary

contrast, taste and preference of family members were considered as factors that encouraged fish consumption13.

Because foods enriched with n-3LCPUFA may play an important role in meeting the recommended intakes for optimal health in Australian children, dietary modelling was conducted to assess the effectiveness of this strategy (Chapter 4). A conceptual approach was designed by replacing the bread, egg, milk and yogurt consumed by children in their actual diet with foods enriched with n-3 LCPUFA that were currently available in supermarkets in Australia. The modelling resulted in a shift of the population distribution of total n-3 LCPUFA intake which resulted in an increase of 15,2 % in the proportion of total children that reached the Adequate Intake reference value. However, the improvement in simulated intakes after modelling remained far below the Suggested Dietary Target level, which was reached by only 6 % of children14.

Since it has been shown in the previous chapters that fish and n-3 enriched foods consumption significantly improved n-3 LCPUFA intake in children, specific food patterns may explain the consumption of these foods. Chapter 5 focused on dietary patterns in relation to n-3 LCPUFA intake. The Principle Component Analysis was used to produce the food patterns. Four food patterns in each gender were identified and were labeled according to the highest positive factor loading. In boys, the food patterns, identified according to the highest factor loading were termed ‘snack foods’, ‘soft drinks’, ‘vegetables’ and ‘pork and meat chops, steak and mince’; while in girls, derived food patterns that emerged were named ‘vegetables’, ‘take aways’, ‘tea, coffee, iced coffee drinks’ and ‘canned meals and soups’. In order to further examine the relationship between the food patterns and n-3 LCPUFA intake, discriminant function analysis was applied. Food patterns that included a high consumption of ‘vegetables’ in boys and ‘take aways’ (meat and fish dishes) in girls were likely to positively influence the total dietary n-3 LCPUFA intake. Both fish and meat influenced n-3 LCPUFA intakes, where the fish was ‘take away fish’ like ‘fish and chips’ and not the oily fish with high n-3 LCPUFA levels; meat influenced n-3 LCPUFA intake because children consumed 8.5 times more meat than fish15.

Given that Australian children did not regularly consume fish or n-3 LCPUFA enriched foods, as was shown in the first two studies of this thesis, the use of the 24-h recall method used in the national Children’s Survey may have led to potential under- 172

Chapter 7. Summary

reporting bias with regard to habitual n-3 LCPUFA intakes. Therefore, in Chapter 6, a FFQ was developed to specifically assess n-3 LCPUFA intake in children. The novel FFQ was developed using data produced from the previous studies reported in Chapters 2, 3, 4 and 5. The FFQ was validated in primary school children. Good correlations between the FFQ and the 7d food dairy were found for EPA, DHA and total n-3 LCPUFA intake, expressed as mg/d. Additionally, Bland-Altman plots showed an acceptable limit of agreement between the two methods. It was therefore concluded that the n-3 LCPUFA FFQ developed in this study provides a useful dietary assessment to estimate dietary n-3 LCPUFA in Australian children. However, the FFQ tended to overestimate the intakes.16

7.2 Future direction and recommendations

This thesis is the first to analyse EPA, DPA and DHA intakes of Australian children, and to identify the major contributors to these nutrients in this age group. It provides information regarding food sources of n-3 LCPUFA in Australian children’s diet that may be targeted through nutrition intervention strategies; therefore it may influence intakes. Furthermore, the approach undertaken in this thesis provides evidence-based information for designing nutrition activities involving children. These include a focus on mediating factors of intake through an approach targeted to parents, recommended consumption of food sources containing n-3 LCPUFA and eating patterns, and also improving dietary assessment tool to monitoring the n-3 LCPUFA intake17.

The data from this thesis supports a number of directions and recommendations for improving dietary n-3 LCPUFA intakes in Australian children. Future research that aims to improve and monitor n-3 LCPUFA intake should consider the following points: (1) frequency of fish and/or seafood consumption, as well as meat, eggs and foods enriched with n-3 LCPUFA products, (2) mediators for consumption of fish and/or seafood and n-3 LCPUFA enriched foods in children, such as parental behaviour in purchasing and preparing fish and n-3 LCPUFA enriched foods for their families; the availability and supply of fish and n-3 LCPUFA enriched foods; and the use of nutrition education strategies to promote consumption of fish and n-3 LCPUFA enriched foods, (3) providing cooking skills to people to improve their confidence in fish meal preparation as well as maintain the health benefits of n-3 LCPUFA in fish, and (4) using validated FFQ to improve the quality of n-3 LCPUFA intake data. 173

Chapter 7. Summary

The findings of this thesis inform potential methods to improve n-3 LCPUFA intakes in Australian children. This information may provide value to the following individuals, professional groups and governmental organizations, as well as institutions.

- General consumers: practical ways to improve n-3 LCPUFA intake. - Health professionals: how to provide dietary advice to improve n-3 LCPUFA intake in Australian children and to maintain the health benefit of n-3 LCPUFA by consuming fish. - Researchers in the area of nutrition and health: some potential points to be followed up or investigated, for example the food enrichment strategies and methodological issues to assess n-3 LCPUFA intakes and public health nutrition promotion. - Food industry: any opportunities for food industries to increase the amount of n-3 LCPUFA substituted in the foods/products and/or opportunities to enhance the enrichment with n-3 LCPUFA to other food products. - Food Standards Australia and New Zealand: the recommendation to meet n-3 LCPUFA by consuming products enriched with n-3 LCPUFA should be evaluated (particularly the number of n-3 LCPUFA enriched foods available in supermarkets). It is due to the fact that these foods are important sources of n-3 LCPUFA in countries that are traditionally low consumers of fish, such as Australia.

The findings of this thesis also complement results/resources developed and produced in a current national project aimed to improve fish consumption in Australia, called the Community Intervention to Improve Seafood Consumption (CIISC)18.

Finally, this thesis contributes valuable new knowledge to the discipline of nutrition and health, particularly with regard to addressing the issues of how to improve and monitor n-3 LCPUFA intake in populations of children, not only in Australia but also in other countries with low intakes of n-3 LCPUFA.

174

Chapter 7. Summary

References

1 Master G, Coles-Rutishauser IH, Webb K, Marks G, Pearse J (2006) A national food and nutrition monitoring and surveillance system: a framework and a business case. Nexus management consulting, Sydney.

2 Hubert HB, Feinleib M, McNamara PM, Castelli WP (1983) Obesity as an independent risk factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham Heart Study. Circulation 67, 968-977.

3 Tounian P, Aggoun Y, Dubern B, Varille V, Guy-Grand B, Sidi D, Girardet JP, Bonnet D (2001) Presence of increased stiffness of the common carotid artery and endothelial dysfunction in severely obese children: a prospective study. Lancet 358, 1400-1404.

4 Venn AJ, Thomson RJ, Schmidt MD, Cleland VJ, Curry BA, Gennat HC, Dwyer T (2007) Overweight and obesity from childhood to adulthood: a follow-up of participants in the 1985 Australian School Health and Fitness Survey. Med J Aust 186, 458-460.

5 Cancer Council Australia, Heart Foundation (2011) High school students graduating to be tomorrow’s cancer and cardiovascular patients. Available at: http://www.heartfoundation.org.au/SiteCollectionDocuments/National-student- survey.pdf, accessed June, 11, 2011.

6 Department of Health and Ageing (2007) User guide - 2007 Australian National Children’s Nutrition and Physical Activity Survey. Available at: http://www.health.gov.au/internet/main/publishing.nsf/Content/phd-nutrition- childrens-survey-userguide, accessed on June, 23, 2010.

7 National Health Medical Research Council (2006) Nutrient reference value for Australia and New Zealand including recommended dietary intakes. Australian Government Department of Health and Ageing. Commonwealth of Australia.

8 Meyer BJ, Kolanu N (2011) Australian children are not consuming enough long- chain omega-3 polyunsaturated fatty acids for optimal health. Nutrition 27, 1136- 1140.

9 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Dietary intake and food sources of omega-3 long chain EPA, DPA and DHA of Australian children. Lipids, In press, accepted manuscript.

10 National Health Foundation of Australia (2008) Position statement, fish, fish oils, n-3 polyunsaturated fatty acids and cardiovascular health. Available at: http://www.heartfoundation.org.au/sitecollectiondocuments/fish-position-statement. pdf, accessed on January, 15, 2011.

11 National Health and Medical Research Council (2013) Australian Dietary Guidelines. Canberra: National Health and Medical Research Council. Available at: http://www.nhmrc.gov.au/guidelines/publications/n55, accessed on March, 25, 2013. 175

Chapter 7. Summary

12 Department of Health (2011) Review of children's healthy eating interventions, 2011. Available at: http://docs.health.vic.gov.au/docs/doc/Full-Review-of-Childrens- Healthy-Eating-Interventions, accessed on January, 25, 2013.

13 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Factors that influence consumption of fish and omega-3 enriched foods: a survey of Australian families with young children. Nutr Diet, doi: 10.1111/1747-0080.12022.

14 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Effect of replacement of bread, egg, milk and yogurt with n-3 enriched for these foods on n-3 LCPUFA intake of Australian children. Br J Nutr, In press, accepted manuscript.

15 Rahmawaty S, Lyons-Wall P, Charlton K, Batterham M, Meyer BJ (2013) Food patterns of Australian children aged 9-13 years in relation to omega-3 long chain polyunsaturated (n-3 LCPUFA) intake. Nutrition, In press, accepted manuscript.

16 Rahmawaty S, Charlton K, Lyons-Wall P, Meyer BJ (2013) Development and validation of a food frequency questionnaire to assess n-3 LCPUFA intake in Australian children aged 9-13 years. Public Health Nutr, Submitted.

17 Department of Health Victoria Australia (2011) Review of children's healthy eating interventions. Available at: http://docs.health.vic.gov.au/docs/doc/Full-Review-of- Childrens-Healthy-Eating-Interventions, accessed on January, 25, 2013

18 McManus A, White J, Hunt W, Storey J, McManus J, Cuesta-Briand B, Golightly A (2011) Community intervention to increase seafood consumption (CIISC). Centre of Excellence for Science Seafood & Health (CESSH), Curtin Health Innovation. Available at: http://cessh.curtin.edu.au/local/docs/ciisc _full.pdf, accessed on March, 20, 2013.

176

Appendices

A p p e n d i c e s

(a) Fish and seafood survey questionnaire (b) Omega-3 LCPUFA FFQ for Australian children aged 9-13 years (c) Food model booklet

177

Appendices

178

Appendices

179

Appendices

180

Appendices

181

Appendices

182

Appendices

183

Appendices

184

Appendices

185

Appendices

186

Appendices

187

Appendices

188

Appendices

189

Appendices

190

Appendices

191

Appendices

192

Appendices

193

Appendices

194

Appendices

195

Appendices

196

Appendices

197

Appendices

198

Appendices

199

Appendices

200

Appendices

201

Appendices

202

Appendices

203

Appendices

204

Appendices

205

Appendices

206

Appendices

207

Appendices

208

Appendices

209

Appendices

210

Appendices

211

Appendices

212

Appendices

213

Appendices

214

Appendices

215

Appendices

216

Appendices

217

Appendices

218

Curriculum Vitae

Setyaningrum Rahmawaty completed her Dipl. III Nutr at the Academic of Nutrition Republic Indonesia, Semarang; Dipl. IV Clin Nutr (DCN) through the Faculty of Medicine, Brawijaya University and master degree (MHSc, specialization in clinical nutrition) at the Faculty of Medicine, Gadjah Mada University where she was awarded with Cum Laude (GPA: 3.82, 4 is the highest). She received a teaching certificate (AKTA IV) from the Faculty of Education Sciences at the University of Muhammadiyah Surakarta with award Cum Laude (GPA: 3.92).

She has worked in academia for a few years at the Department of Nutrition, University of Muhammadiyah Surakarta in Indonesia before undertaking her PhD at the University of Wollongong Australia. She is interested in diets and their relation to metabolic diseases, development of nutrition education materials and the use of local foods for nutrition therapy. She has received research grants prior doing her PhD e.g. multiyear research grant from the Ministry of Education, Indonesia (2007 and 2008), modelling teaching learning system grant addressed to improve dietetic competences and lifelong learning (Ministry of Education, Indonesia, 2006).

During doing her PhD, she has produced a number of journal articles published in: 1) Lipids, 2) Nutrition and Dietetics, 3) Nutrition and, 4) Journal of Public Health Nutrition where she was the main author. She has contributed in a review paper published in the Journal of Nutritional Biochemistry (art work designs about the mechanism of n-3 LCPUFA in brain).

219