Eating behaviour is associated with eating frequency and food consumption in 6-8 year-old children: The Physical Activity and Nutrition in Children (PANIC) study

DSpace https://erepo.uef.fi

Rinnakkaistallenteet Terveystieteiden tiedekunta

2017

Eating behaviour is associated with eating frequency and food

consumption in 6-8 year-old children:

The Physical Activity and Nutrition in Children (PANIC) study

Jalkanen H

info:eu-repo/semantics/article

info:eu-repo/semantics/acceptedVersion

© Elsevier B.V

CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/

http://dx.doi.org/10.1016/j.appet.2017.03.011

https://erepo.uef.fi/handle/123456789/4251

Downloaded from University of Eastern Finland's eRepository

Eating behaviour is associated with eating frequency and food consumption in 6–8 year-old children: The Physical Activity and Nutrition in Children (PANIC) study H. Jalkanen, V. Lindi, U. Schwab, S. Kiiskinen, T. Venäläinen, L. Karhunen, T.A.

Lakka, A.M. Eloranta

PII: S0195-6663(17)30384-7

DOI: 10.1016/j.appet.2017.03.011 Reference: APPET 3372

To appear in: Appetite Received Date: 29 August 2016 Revised Date: 13 January 2017 Accepted Date: 9 March 2017

Please cite this article as: Jalkanen H., Lindi V., Schwab U., Kiiskinen S., Venäläinen T., Karhunen L., Lakka T.A. & Eloranta A.M., Eating behaviour is associated with eating frequency and food consumption in 6–8 year-old children: The Physical Activity and Nutrition in Children (PANIC) study, Appetite (2017), doi: 10.1016/j.appet.2017.03.011.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

M AN US CR IP T

AC CE PT ED

1 Eating behaviour is associated with eating frequency and food consumption in 6-8 year- 1

old children: the Physical Activity and Nutrition in Children (PANIC) Study 2

Jalkanen H¹, Lindi V¹, Schwab U^2,3, Kiiskinen S¹, Venäläinen T^1,2, Karhunen L^2,3, Lakka 3

TA^1,4,5, Eloranta AM¹ 4

5

1Institute of Biomedicine, Physiology, School of Medicine, University of Eastern Finland, 6

Kuopio, Finland 7

2Institute of Public Health and Clinical Nutrition, Clinical Nutrition, School of Medicine, 8

University of Eastern Finland, Kuopio, Finland 9

3Institute of Clinical Medicine, Internal Medicine, Kuopio University Hospital, Kuopio, 10

Finland 11

4Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital, 12

Kuopio, Finland 13

5Kuopio Research Institute of Exercise Medicine, Kuopio, Finland 14

15

Corresponding author:

16

Henna Jalkanen, M.Sc. (Nutrition) 17

University of Eastern Finland, Institute of Biomedicine, Physiology 18

PO Box 1627 19

Fin-70211 Kuopio, Finland.

20

Tel: +358400496686. E-mail: henna.jalkanen@uef.fi 21

22

23

M AN US CR IP T

AC CE PT ED

2 Abstract

24

The association between eating behaviour and dietary factors has been studied narrowly in 25

children. Therefore, we investigated whether eating frequency and food consumption are 26

influenced by eating behaviour in a population sample of 406 children aged 6-8 years. We 27

assessed features of eating behaviour by the Children's Eating Behaviour Questionnaire and 28

dietary factors by a 4-day food record. The results showed that enjoyment of food was directly 29

associated with a number of main meals (p=0.041) and consumption of vegetables (p=0.041), 30

cheese (p=0.005), and meat (p=0.002). Food responsiveness was directly associated with 31

consumption of fruit and berries (p=0.013) and meat (p=0.016). Desire to drink was directly 32

associated with consumption of fat-containing milk (p=0.002) and inversely associated with 33

consumption of skimmed milk (p=0.001). Food fussiness was inversely associated with a 34

number of main meals (p=0.013) and consumption of vegetables (p<0.001), cheese 35

(p=0.001), and meat (p=0.027). Satiety responsiveness was inversely associated with 36

consumption of vegetables (p=0.031), cheese (p=0.010), and meat (p<0.001) and directly 37

associated with consumption of candies and chocolate (p=0.026). Slowness in eating was 38

inversely associated with consumption of meat (p=0.018). Where sex differences existed the 39

associations tended to be observed mostly in girls but not in boys. Our study shows that 40

enjoyment of food and food responsiveness are directly associated with consumption of 41

protein-rich foods and vegetables, fruit and berries, whereas food fussiness and satiety 42

responsiveness are inversely associated with consumption of these foods. Assessment of 43

eating behaviour can help in identifying children with various dietary needs.

44

Keywords: Children; Eating behaviour; Appetitive traits; Eating frequency; Food 45

consumption; Diet 46

47

M AN US CR IP T

AC CE PT ED

3 Introduction

48

School-aged children in the developed countries commonly eat too much foods containing 49

lots of sucrose, saturated fat, and salt, such as sugar-sweetened beverages, candies, and meat, 50

and too little foods high in vitamins, minerals, fibre, and other nutrients that are essential for 51

growth, development, and health, such as vegetables, fruit, berries, high-fibre grain products, 52

and fish (Diethelm et al., 2012; Elmadfa et al., 2009; Eloranta et al., 2011; Hoppu et al., 53

2010; Kyttala et al., 2010; Lambert et al., 2004). School-aged children also tend to skip meals 54

and this habit has been found to increase the risk of becoming overweight in childhood 55

(Eloranta et al., 2012; Jaaskelainen et al., 2013). One crucial way to facilitate normal growth 56

and development and reduce the risk of health problems originating in childhood would be to 57

improve the eating habits of school-aged children.

58

Previous studies have found that certain features of eating behaviour, such as food approach 59

eating behaviour sub-scales enjoyment of food and food responsiveness, have been related to 60

weight (Croker et al., 2011; Webber et al., 2009), body fat percentage (Eloranta et al., 2012), 61

and obesity in children (McCarthy et al., 2015). In addition, food-avoidant eating behaviour 62

sub-scale food fussiness has been associated with the risk of inadequate energy and nutrient 63

intake (Galloway et al., 2005). These previous findings emphasise the important link between 64

eating behaviour and health aspects in children. Some studies have also found an association 65

between eating behaviour and food preferences in children. For example, food fussiness and 66

satiety responsiveness have been associated with a lower liking of vegetables and fruit (Fildes 67

et al., 2015). However, rather few studies in children have investigated the associations 68

between eating behaviour and actual food consumption (Carnell et al., 2016; Cooke et al., 69

2004; Dubois et al., 2007; Galloway et al., 2005; Rodenburg et al., 2012) and meal frequency 70

(Syrad et al., 2016). In these studies food responsiveness has been associated with a higher 71

meal frequency (Syrad et al., 2016) and a higher consumption of vegetables and fruit (Carnell 72

M AN US CR IP T

AC CE PT ED

4 et al., 2016; Cooke et al., 2004), whereas so-called picky or fussy eating has been associated 73

with a lower consumption of these foods (Dubois et al., 2007; Galloway et al., 2005).

74

According to previous studies, eating behaviour does not differ between girls and boys at 75

preschool age (Svensson et al., 2011) or if there are differences, they are small when children 76

are aged 2 to 7 years (Wardle et al., 2001). However, boys have been reported to score higher 77

in food fussiness and emotional overeating and lower in enjoyment of food than girls when 78

they reach primary school age (Sleddens et al., 2008). In addition, a systematic literature 79

review concluded that boys were more likely to eat breakfast than girls (Currie et al., 2012) 80

but that girls consumed more vegetables and fruit than boys (Rasmussen et al., 2006).

81

Therefore, there may be some differences in factors affecting the food selection between boys 82

and girls that only begin to manifest later on in childhood. However, there are no earlier 83

studies investigating differences in the associations of eating behaviour with eating frequency 84

and food consumption between primary school aged boys and girls.

85

It is important to determine which factors may influence food intake in boys and girls in order 86

to enhance their diet quality, health and growth. We therefore investigated the association of 87

eating behaviour with eating frequency and food consumption among 6-8-year-old primary- 88

school children and whether there are sex differences in these associations.

89

90

Methods 91

Study population and study design 92

This is a part of the Physical Activity and Nutrition in Children (PANIC) study which is an 93

ongoing physical activity and dietary intervention in a population sample of children 94

(Eloranta et al., 2012). A total of 736 children aged 6-8 years who started first grade in 16 95

M AN US CR IP T

AC CE PT ED

5 primary schools of the city of Kuopio, Finland, in 2007–2009, were invited to participate in 96

the study via letters forwarded to their parents from the schools, with 512 (70 %) 97

participating in the baseline evaluation. According to the school health examination data, the 98

participants did not differ in age, sex, distribution, or body mass index standard deviation 99

score (BMI-SDS) from all children who started the first grade in the primary schools of 100

Kuopio during the years 2007–2009. The final study population comprised of 406 children 101

(204 girls and 202 boys) who had complete data on food consumption and eating behaviour.

102

The participants did not differ in terms of age, sex distribution, or BMI-SDS from those 103

children (n=106) who were excluded from the analyses due to missing data.

104

The PANIC study protocol was approved by the Research Ethics Committee of the Hospital 105

District of Northern Savo (Kuopio, Finland). All the children and their parents gave their 106

written informed consent for participation in the study.

107

108

Assessment of body size and composition 109

Anthropometric measures were assessed at the beginning of the study. Body height (cm) was 110

measured to an accuracy of 0.1 cm using a wall-mounted stadiometer. Body weight (kg) was 111

measured to an accuracy of 0.1 kg after overnight fasting and with an empty-bladder with an 112

InBody 720 bioelectrical impedance device (Biospace, Seoul, Korea). Body mass index 113

(BMI) was calculated by dividing mean body weight in kg by mean body height in meters 114

squared (kg m¯ ²). BMI-SDS was calculated based on Finnish references (Saari et al., 2011) 115

and overweight and obesity were defined using international cut-off values (Cole et al., 116

2000).

117

118

Assessment of eating behaviour 119

M AN US CR IP T

AC CE PT ED

6 We assessed eating behaviour using the Children’s Eating Behaviour Questionnaire (CEBQ) 120

(Carnell & Wardle, 2007; Wardle et al., 2001) translated into Finnish using the forward- 121

backward translation method (Seppänen, 2005). On the first study visit, parents were given 122

instructions on how to complete at home the 35-item questionnaire on behalf of their child.

123

Parents returned the questionnaire on the second visit to the researchers who checked and 124

filled the missing information together with parents.

125

CEBQ was used to assess eight different features of eating behaviour: enjoyment of food, 126

food responsiveness, emotional overeating, desire to drink, food fussiness, satiety 127

responsiveness, emotional undereating, and slowness in eating with the eight corresponding 128

subscales. Each subscale consists of 3–6 statements for example: “My child loves food”

129

(enjoyment of food), “My child is always asking for food” (food responsiveness), “My child 130

eats more when worried” (emotional overeating), “My child is always asking for a drink”

131

(desire to drink), “My child refuses new foods at first” (food fussiness), “My child gets full 132

up easily” (satiety responsiveness), “My child eats less when angry” (emotional 133

undereating), and “My child eats slowly” (slowness in eating), the response options given as 134

five-point Likert-scales (never = 1, rarely = 2, sometimes = 3, often = 4, and always = 5). The 135

mean of each subscale was used in the analyses. The greater the mean value the greater the 136

prevalence of that feature in that child’s eating behaviour.

137

Internal reliability coefficients (Cronbach’s alphas) were calculated for each eating behaviour 138

subscale, and furthermore, item to total correlations were calculated for all statements of the 139

subscales. Good internal consistency was found in seven of the eight studied subscales, with 140

values of Cronbach’s alpha ranging from 0.765 to 0.896 (Table 1). Only the subscale 141

emotional undereating exhibited a low internal consistency. In that subscale, one of the 142

statements (“My child eats more when s/he is happy”) showed a low item to total correlation, 143

whereas the item to total correlations for the other statements were within acceptable ranges 144

M AN US CR IP T

AC CE PT ED

7 (r = 0.382 to 0.450). When we calculated the subscale emotional undereating after excluding 145

the divergent statement, the subscale showed good internal consistency (Table 1). Therefore, 146

we used the modified subscale in the subsequent analyses.

147

148

149

M AN US CR IP T

AC CE PT ED

8 Table 1. Internal reliability coefficients (Cronbach’s alphas) of subscales and item to total correlations of each statement of Children’s Eating 150

Behaviour Questionnaire.

151

Item Cronbach's alpha Item to total correlation

Enjoyment of food 0.785

My child loves food 0.707

My child is interested in food 0.714

My child looks forward to mealtimes 0.552

My child enjoys eating 0.618

Food responsiveness 0.785

My child is always asking for food 0.527

If allowed to, my child would eat too much 0.608

Given the choice, my child would eat most of the time 0.672

Even if my child is full up s/he finds room to eat his/her favourite food 0.453 If given the chance, my child would always have food in his/her mouth 0.644

Emotional overeating 0.771

My child eats more when worried 0.649

My child eats more when annoyed 0.658

My child eats more when anxious 0.628

My child eats more when s/he has nothing else to do 0.500

Desire to drink 0.777

My child is always asking for a drink 0.645

If given the chance, my child would drink continuously throughout the day 0.683

If given the chance, my child would always be having a drink 0.778

Food fussiness 0.896

My child refuses new foods at first 0.781

My child enjoys tasting new foods 0.770

My child enjoys a wide variety of foods 0.717

My child is difficult to please with meals 0.578

M AN US CR IP T

AC CE PT ED

9 My child is interested in tasting food s/he hasn’t tasted before 0.779

My child decides that s/he doesn’t like a food, even without tasting it 0.710

Satiety responsiveness 0.765

My child has a big appetite 0.505

My child leaves food on his/her plate at the end of a meal 0.633

My child gets full before his/her meal is finished 0.553

My child gets full up easily 0.503

My child cannot eat a meal if s/he has had a snack just before 0.489

Emotional undereating^a 0.176 (0.757)

My child eats less when angry 0.450 (0.658)

My child eats less when s/he is tired 0.404 (0.534)

My child eats more when s/he is happy -0.549 ( - )

My child eats less when upset 0.382 (0.583)

Slowness in eating 0.805

My child finishes his/her meal quickly 0.596

My child eats slowly 0.743

My child takes more than 30 minutes to finish a meal 0.581

My child eats more and more slowly during the course of a meal 0.568

a The Cronbach’s alpha and items to total correlations of emotional undereating after excluding the divergent statement “My child eats more 152

when s/he is happy” are reported in brackets.

153

154

155

156

M AN US CR IP T

AC CE PT ED

10 Assessment of eating frequency and food consumption

157

We assessed eating frequency and food consumption using 4-day food records. On their first 158

study visit, the family were provided with instructions by a clinical nutritionist on how to fill 159

in the food diary. Families were asked to record detailed descriptions and portion sizes of all 160

the foods and beverages that their child consumed on 4 consecutive, pre-determined days, 161

including 2 weekend days and 2 weekdays, at home, at school, in afternoon care, and 162

elsewhere outside home. Moreover, the details of food preparation methods, such as added 163

fats and cooking fat, were enquired. Details regarding the served food and preparation style in 164

the schools and afterschool clubs were gathered from the catering company that provided the 165

food to the schools. The meals were defined according to the reported time and the type of 166

food taking into account the whole meal pattern of each child individually. Breakfast, lunch, 167

and dinner were classified as main meals, and all eating and drinking occasions between the 168

main meals were classified as snacks.

169

A clinical nutritionist checked the returned food records together with the family and filled 170

in any missing information on their second visit. The food consumption was calculated using 171

Micro-Nutrica® dietary analysis software, version 2.5 (The Social Insurance Institution, 172

Finland). Micro-Nutrica® software is updated regularly by adding new food items and 173

products with precise nutritional contents provided by producers. In addition, a clinical 174

nutritionist divided the foods consumed and listed in the software into 20 food groups each of 175

which included foods of similar nutrient compositions and types of consumption (Table 2).

176

We estimated that total energy intake had been underestimated in 27% of the girls and 24%

177

of the boys when comparing it with energy expenditure estimated by basal metabolic rate 178

calculated using Schofield’s equation and using the cut-off values for under-reporting 179

suggested by Torun et al. (Torun et al., 1996).

180

181

M AN US CR IP T

AC CE PT ED

11 Statistical methods

182

The SPSS statistical software Version 21 (IBM Corp, Armonk, NY, USA) was used to 183

perform statistical analyses. Means and standard deviations of basic characteristics subscale 184

scores of eating behaviour, eating frequency, and food consumption were calculated. T test 185

for independent samples and Pearson’s χ2-test were used to compare the basic characteristics, 186

subscale scores of eating behaviour, eating frequency, and food consumption between girls 187

and boys. The associations of subscale scores of eating behaviour with eating frequency and 188

food consumption were analysed using linear regression models adjusted for age and sex. We 189

also analysed the associations excluding under-reporters. Because the directions and 190

magnitudes of the associations remained similar after this exclusion (data not shown), we 191

only report the associations in the whole study sample to achieve higher statistical power. We 192

used the general linear models to test if there was an interaction between sex and eating 193

behaviour subscales on eating frequency and food consumption. If there was a statistically 194

significant association, linear regression models on the associations of eating behaviour with 195

eating frequency and food consumption were additionally calculated for girls and boys 196

separately. Associations with a p-value of < 0.05 were considered statistically significant. We 197

also used the Bonferroni correction for multiple testing in analyses because of the large 198

number of analyses performed. The threshold of statistical significance with the Bonferroni 199

correction was computed as the p-value of 0.05 divided by the number of variables used in 200

each analysis.

201

202

Results 203

Characteristics 204

M AN US CR IP T

AC CE PT ED

12 Demographic and anthropometric characteristics, eating behaviour and food consumption in 205

the study population are presented in Table 2. Boys were taller, had lower scores in satiety 206

responsiveness, emotional undereating, and slowness in eating, and consumed more sour 207

milk products (fat ≥ 1%), meat, and candies and chocolate than girls.

208

209

210

M AN US CR IP T

AC CE PT ED

13 Table 2. Characteristics, eating behaviour, eating frequency, and food consumption in children.

211

All children (n=406) Mean ± SD

Girls (n=204) Mean ± SD

Boys (n=202)

Mean ± SD p-value *

Characteristics

Age (years) 7.6 ± 0.4 7.6 ± 0.4 7.7 ± 0.4 0.279

Height (cm) 128.9 ± 5.6 128.2 ± 5.7 129.7 ± 5.5 0.008

Weight (kg)^a 27.0 ± 5.0 26.7 ± 5.2 27.3 ± 4.9 0.192

Body mass index standard deviation score^b -0.17 ± 1.1 -0.17 ± 1.1 -0.17 ± 1.1 0.969 Weight status^c

Normal weight (%, n) 87.2 (353) 85.3 (174) 89.1 (179)

Overweight (%, n) 8.6 (35) 10.8 (22) 6.5 (13)

Obesity (%, n) 4.2 (17) 3.9 (8) 4.5 (9) 0.298

Eating behaviour subscales

Enjoyment of food 3.2 ± 0.7 3.2 ± 0.6 3.2 ± 0.7 0.210

Food responsiveness 1.7 ± 0.4 1.7 ± 0.5 1.7 ± 0.6 0.885

Emotional overeating 1.4 ± 0.5 1.4 ± 0.5 1.5 ± 0.5 0.184

Desire to drink 1.8 ± 0.7 1.8 ± 0.7 1.8 ± 0.7 0.843

Food fussiness 2.9 ± 0.8 2.9 ± 0.8 2.8 ± 0.7 0.510

Satiety responsiveness 3.0 ± 0.6 3.1 ± 0.6 2.9 ± 0.6 <0.001

Emotional undereating 2.9 ± 0.6 2.9 ± 0.6 2.8 ± 0.5 0.010

Slowness in eating 2.7 ± 0.7 2.9 ± 0.7 2.5 ± 0.7 <0.001

Eating frequency (number per day)

Main meals 2.8 ± 0.3 2.7 ± 0 3 2.8 ± 0.3 0.098

Snacks 2.7 ± 0.9 2.8 ± 0.9 2.7 ± 0.9 0.710

Food consumption (g/day)

High-fibre grain products^d 63.8 ± 39.5 60.9 ± 36.1 66.8 ± 42.5 0.134

Low-fibre grain products^d 112.3 ± 52.1 108.1 ± 44.9 116.5 ± 58.3 0.106

Vegetables^e 102.1 ± 59.4 103.8 ± 59.7 100.3 ± 59.2 0.563

Potato 76.4 ± 41.6 73.4 ± 41.5 79.5 ± 41.6 0.143

M AN US CR IP T

AC CE PT ED

14

Fruit and berries 108.6 ± 83.8 114.7 ± 84.5 102.4 ± 82.9 0.140

Butter and butter-oil mixture 5.9 ± 7.1 5.3 ± 6.1 6.4 ± 8.0 0.124

Vegetable-oil based margarine (fat 60-80%) 7.3 ± 8.0 7.3 ± 7.9 7.2 ± 8.3 0.842 Low-fat vegetable-oil based margarine

(fat <60%) 3.9 ± 6.9 3.3 ± 5.6 4.5 ± 8.0 0.062

Milk (fat ≥1 %) 190.8 ± 226.3 183.1 ± 214.8 198.5 ± 237.6 0.494

Skimmed milk (fat <1%) 386.9 ± 294.2 358.8 ± 277.9 415.3 ± 308.0 0.053

Sour milk products (fat ≥1%)^f 85.5 ± 80.2 77.7 ± 67.5 93.4 ± 90.7 0.049

Low-fat sour milk products (fat <1%)^f 18.6 ± 53.2 18.7 ± 51.9 18.4 ± 54.6 0.958

Cheese 15.6 ± 15.7 14.8 ± 14.1 16.4 ± 17.1 0.301

Ice cream and pudding 26.7 ± 32.1 27.0 ± 32.2 26.4 ± 32.1 0.863

Meat^g 94.6 ± 41.7 85.9 ± 38.6 103.7 ± 42.8 <0.001

Fish 16.0 ± 21.2 14.0 ± 19.2 18.1 ± 23.0 0.053

Fruit juices 36.7 ± 67.9 32.4 ± 50.0 41.1 ± 82.0 0.195

Sugar-sweetened beverages^h 134.5 ± 126.0 124.3 ± 118.5 144.8 ± 133.1 0.103

Candies and chocolate 30.5 ± 26.9 27.4 ± 25.0 33.7 ± 28.5 0.018

Spices and table sauces 12.9 ± 19.3 12.7 ± 15.8 13.1 ± 22.4 0.846

SD, standard deviation. *Sex differences were assessed by the t-test for independent samples and the Pearson’s χ²-test. P-values <0.05 are bolded.

a n=405 (203 girls, 202 boys)

b Based on Finnish reference values (Saari et al., 2011)

c Weight status defined using international cut-offs (Cole et al., 2000)

d Includes bread, cereal, porridge, flour, pasta, and rice.

e Includes vegetables, roots, beans, and mushrooms excluding potato.

f Includes yoghurt, curdled milk, sour milk, and quark.

g Includes pork, beef, lamb, reindeer, game meat, and sausages.

h Includes carbonated and non-carbonated beverages.

212

M AN US CR IP T

AC CE PT ED

15 Associations of enjoyment of food with eating frequency and food consumption

213

A higher score in the subscale enjoyment of food was associated with a higher number of 214

meals per day in the whole study population (Table 3) and in girls (β = 0.206, p = 0.003) but 215

not in boys (p = 0.037 for interaction).

216

A higher score in enjoyment of food was associated with a higher consumption of vegetables 217

in the whole study population (Table 3) and in girls (β = 0.177, p = 0.011) but not in boys (p 218

= 0.088 for interaction). A higher score in enjoyment of food was also associated with a 219

higher consumption of cheese in the whole study population (Table 3). In addition, a higher 220

score in enjoyment of food was associated with a higher consumption of meat in the whole 221

study population (Table 3) and in girls (β = 0.223, p = 0.001) but not in boys (p = 0.210 for 222

interaction).

223

224

225

M AN US CR IP T

AC CE PT ED

16 Table 3. Associations of enjoyment of food, food responsiveness, emotional overeating, and desire to drink with eating frequency and food 226

consumption.

227

Enjoyment of food Food responsiveness Emotional overeating Desire to drink

β p-value β p-value β p-value β p-value

Eating frequency (number/day)

Main meals 0.101 0.041 -0.070 0.160 -0.083 0.093 -0.067 0.174

Snacks -0.011 0.820 0.060 0.230 0.095 0.059 0.108 0.030

Food consumption (g/day)

High-fibre grain products^a 0.064 0.197 0.081 0.102 -0.158 0.002 -0.098 0.049

Low-fibre grain products^a 0.083 0.096 <0.001 0.996 0.096 0.054 0.011 0.824

Vegetables^b 0.102 0.041 0.046 0.358 0.036 0.475 0.020 0.695

Potato 0.096 0.053 0.075 0.134 0.072 0.151 0.029 0.567

Fruit and berries 0.010 0.835 0.122 0.013 0.055 0.272 -0.015 0.758

Butter and butter-oil mixture 0.026 0.607 -0.047 0.349 -0.030 0.548 0.029 0.564

Vegetable-oil based margarine (fat 60-80 %)

-0.006 0.908 0.006 0.900 -0.005 0.923 0.002 0.967

Low-fat vegetable-oil based margarine (fat <60 %)

-0.039 0.430 0.069 0.167 -0.024 0.631 0.046 0.358

Milk (fat ≥1 %) 0.046 0.357 -0.029 0.567 -0.023 0.649 0.153 0.002

Skimmed milk (fat <1 %) -0.088 0.077 0.007 0.892 -0.028 0.581 -0.163 0.001

Sour milk products (fat ≥1 %)^c -0.014 0.773 -0.040 0.418 -0.016 0.746 0.036 0.461

Low-fat sour milk 0.011 0.829 -0.007 0.889 -0.050 0.316 -0.015 0.758

products (fat <1 %)^c

Cheese 0.141 0.005 0.041 0.414 -0.021 0.674 0.020 0.695

Ice cream and pudding <0.001 0.997 0.015 0.766 0.028 0.575 0.109 0.026

Meat ^d 0.152 0.002 0.117 0.016 0.090 0.065 0.053 0.278

Fish 0.034 0.494 -0.062 0.213 0.017 0.740 -0.065 0.193

Fruit juices 0.006 0.905 -0.056 0.266 -0.061 0.224 0.021 0.670

M AN US CR IP T

AC CE PT ED

17

Sugar-sweetened beverages^e -0.086 0.080 0.019 0.701 0.006 0.903 0.004 0.934

Candies and chocolate -0.059 0.237 0.006 0.901 0.003 0.945 -0.044 0.378

Spices and table sauces 0.079 0.112 0.017 0.739 -0.112 0.023 -0.080 0.106

Standardised regression coefficients (β) are from linear regression models adjusted for age and sex. The threshold of statistical significance with Bonferroni correction is 0.002. P-values <0.05 are bolded.

a Includes bread, cereal, porridge, flour, pasta, and rice.

b Includes vegetables, roots, beans, and mushrooms excluding potato.

c Includes yoghurt, curdled milk, sour milk, and quark.

d Includes pork, beef, lamb, reindeer, game meat, and sausages.

e Includes carbonated and non-carbonated beverages.

M AN US CR IP T

AC CE PT ED

18 Associations of food responsiveness with eating frequency and food consumption

228

Food responsiveness was not associated with eating frequency.

229

A higher score in the subscale food responsiveness was associated with a higher consumption 230

of fruit and berries in the whole study population (Table 3) and in girls (β = 0.249, p < 0.001) 231

but not in boys (p = 0.013 for interaction). A higher score in food responsiveness was also 232

associated with a higher consumption of meat in the whole study population (Table 3).

233

234

Associations of emotional overeating with eating frequency and food consumption 235

Emotional overeating was not associated with eating frequency.

236

A higher score in the subscale emotional overeating was associated with a lower 237

consumption of high-fibre grain products in the whole study population (Table 3), in girls (β 238

= -0.174, p = 0.013), and in boys (β = -0.141, p = 0.048) (p = 0.870 for interaction). In 239

addition, a higher score in emotional overeating was associated with a lower consumption of 240

spices and table sauces in the whole study population (Table 3) and in girls (β = -0.181 and p 241

= 0.009) but not in boys (p = 0.401 for interaction).

242

243

Associations of desire to drink with eating frequency and food consumption 244

A higher score in the subscale desire to drink was associated with a higher number of snacks 245

per day in the whole study population (Table 3) and in girls (β = 0.193, p = 0.006) but not in 246

boys (p = 0.058 for interaction).

247

A higher score in desire to drink was associated with a lower consumption of high-fibre grain 248

products (Table 3) in the whole study population. In addition, a higher score in desire to drink 249

was associated with a higher consumption of milk (fat ≥ 1%) in the whole study population 250

(Table 3) and in girls (β = 0.206, p = 0.003) but not in boys (p = 0.416 for interaction). A 251

M AN US CR IP T

AC CE PT ED

19 higher score in desire to drink was associated with a lower consumption of skimmed milk in 252

the whole study population (Table 3) and in girls (β = -0.206, p = 0.003) but not in boys (p = 253

0.542 for interaction). A higher score in desire to drink was associated with a higher 254

consumption of ice cream and pudding in the whole study population (Table 3) and in girls (β 255

= 0.165, p = 0.017) but not in boys (p = 0.224 for interaction).

256

257

The associations of food fussiness with eating frequency and food consumption 258

A higher score in the subscale food fussiness was associated with a lower number of main 259

meals per day in the whole study population (Table 4) and in girls (β = -0.230, p = 0.001) but 260

not in boys (p = 0.033 for interaction).

261

A higher score in food fussiness was associated with a lower consumption of vegetables in 262

the whole study population (Table 4), in girls (β = -0.268, p < 0.001), and in boys (β = -0.183, 263

p = 0.008) (p = 0.425 for interaction). A higher score in food fussiness was also associated 264

with a lower consumption of cheese in the whole study population (Table 4) and in boys (β = 265

-0.195, p = 0.006) but not in girls (p = 0.306 for interaction). A higher score in food fussiness 266

was associated with a lower consumption of meat in the whole study population (Table 4) 267

and in girls (β = -0.193, p = 0.006) but not in boys (p = 0.143 for interaction). In addition, a 268

higher score in food fussiness was associated with a lower consumption of spices and table 269

sauces in the whole study population (Table 4) and in boys (β = -0.194, p = 0.006) but not in 270

girls (p = 0.098 for interaction).

271

272

M AN US CR IP T

AC CE PT ED

20 Table 4. Associations of food fussiness, satiety responsiveness, emotional undereating, and slowness in eating with eating frequency and food 273

consumption.

274

Food fussiness Satiety responsiveness Emotional undereating Slowness in eating

β p-value β p-value β p-value β p-value

Eating frequency (number/day)

Main meals -0.122 0.013 -0.041 0.417 -0.025 0.619 -0.039 0.451

Snacks 0.046 0.361 0.032 0.533 0.056 0.262 0.016 0.755

Food consumption (g/day)

High-fibre grain products^a -0.083 0.093 -0.193 <0.001 -0.044 0.383 -0.063 0.225

Low-fibre grain products^a -0.014 0.784 -0.123 0.016 -0.019 0.711 -0.085 0.098

Vegetables^b -0.233 <0.001 -0.110 0.031 0.050 0.316 0.006 0.910

Potato -0.038 0.439 -0.097 0.058 -0.051 0.314 -0.030 0.563

Fruit and berries -0.069 0.163 -0.034 0.506 -0.041 0.408 0.048 0.346

Butter and butter-oil mixture -0.033 0.503 -0.064 0.211 0.083 0.099 -0.010 0.853 Vegetable-oil based

margarine (fat 60-80 %)

0.051 0.307 -0.006 0.912 -0.013 0.795 0.011 0.835

Low-fat vegetable-oil based margarine (fat <60 %)

0.030 0.552 -0.037 0.469 -0.023 0.647 0.039 0.448

Milk (fat ≥1 %) -0.008 0.870 0.057 0.261 0.016 0.755 0.099 0.056

Skimmed milk (fat <1 %) 0.024 0.625 0.023 0.646 -0.069 0.164 -0.010 0.847

Sour milk products (fat ≥1 %)^c -0.083 0.093 -0.059 0.245 -0.035 0.487 -0.006 0.911

Low-fat sour milk -0.017 0.740 0.022 0.660 0.044 0.384 0.046 0.377

products (fat <1 %)^c

Cheese -0.167 0.001 -0.130 0.010 -0.046 0.360 -0.007 0.899

Ice cream and pudding 0.079 0.107 0.052 0.303 -0.006 0.905 0.049 0.333

Meat^d -0.107 0.027 -0.246 <0.001 -0.087 0.074 -0.119 0.018

Fish -0.095 0.056 0.039 0.447 0.060 0.233 -0.010 0.841

Fruit juices 0.097 0.052 0.056 0.276 0.060 0.229 -0.057 0.269

M AN US CR IP T

AC CE PT ED

21

Sugar-sweetened beverages^e 0.086 0.081 -0.045 0.377 -0.013 0.796 0.003 0.946

Candies and chocolate 0.011 0.826 0.113 0.026 0.057 0.253 0.015 0.776

Spices and table sauces -0.146 0.003 0.078 0.127 -0.041 0.410 0.006 0.901

Standardised regression coefficients (β) are from linear regression models adjusted for age and sex. The threshold of statistical significance with Bonferroni correction is 0.002. P-values <0.05 are bolded.

a Includes bread, cereal, porridge, flour, pasta, and rice.

b Includes vegetables, roots, beans, and mushrooms excluding potato.

c Includes yoghurt, curdled milk, sour milk, and quark.

d Includes pork, beef, lamb, reindeer, game meat, and sausages.

e Includes carbonated and non-carbonated beverages.

M AN US CR IP T

AC CE PT ED

22 The associations of satiety responsiveness with eating frequency and food consumption 275

Satiety responsiveness was not associated with eating frequency.

276

A higher score in the subscale satiety responsiveness was associated with a lower 277

consumption of high-fibre grain products in the whole study population (Table 4), in girls (β 278

= -0.150, p = 0.033), and in boys (β = -0.242, p = 0.001) (p = 0.238 for interaction). In 279

addition, a higher score in satiety responsiveness was associated with a lower consumption of 280

low-fibre grain products in the whole study population (Table 4) and in boys (β = -0.163, p = 281

0.023) but not in girls (p = 0.134 for interaction). A higher score in satiety responsiveness 282

was also associated with a lower consumption of vegetables in the whole study population 283

(Table 4) and in girls (β = -0.146, p = 0.036) but not in boys (p = 0.468 for interaction). A 284

higher score in satiety responsiveness was associated with a lower consumption of cheese in 285

the whole study population (Table 4). A higher score in satiety responsiveness was associated 286

with a lower consumption of meat in the whole study population (Table 4), in girls (β = - 287

0.272, p < 0.001), and in boys (β = -0.225, p = 0.002) (p = 0.908 for interaction). A higher 288

score in satiety responsiveness was associated with a higher consumption of candies and 289

chocolate in the whole study population (Table 4) and in boys (β = 0.154, p = 0.032) but not 290

in girls (p = 0.156 for interaction).

291

292

The associations of emotional undereating with eating frequency and food consumption 293

Emotional undereating was not associated with eating frequency or the individual aspects of 294

food consumption.

295

296

The associations of slowness in eating with eating frequency and food consumption 297

M AN US CR IP T

AC CE PT ED

23 Slowness in eating was not associated with eating frequency.

298

A higher score in slowness in eating was associated with a lower consumption of meat in the 299

whole study population (Table 4) and in girls (β = -0.150, p = 0.032) but not in boys (β = - 300

0.087, p = 0.216) (p = 0.635 for interaction).

301

302

Discussion 303

The results of this study revealed that eating behaviour was associated with dietary factors in 304

a population sample of 6-8 years old children. In particular, the higher scores in enjoyment of 305

food and food responsiveness subscales were associated with a higher consumption of 306

protein-rich foods, such as meat, and of vegetables, fruit, and berries, whereas higher scores 307

in food fussiness and satiety responsiveness subscales were associated with a lower 308

consumption of these foods. When sex differences existed in the associations, the 309

associations tended to be observed mostly in girls but not in boys.

310

It has been shown that enjoyment of food is positively related to the consumption of fruit 311

(Rodenburg et al., 2012) and a liking for vegetables and fruit (Fildes et al., 2015). In addition, 312

Enjoyment of food has been associated with a higher food variety in children (Russell &

313

Worsley, 2016). We found that a higher score in enjoyment of food was positively associated 314

with the consumption of vegetables and that food responsiveness was positively associated 315

with the consumption of fruit and berries. Food responsiveness and enjoyment of food were 316

also positively associated with meal type foods, such as meat, and enjoyment of food was 317

positively associated with the number of main meals per day. Similarly, food responsiveness 318

has been associated with more eating occasions in one previous study (Syrad et al., 2016).

319

This may indicate that children with these eating behaviour tendencies are more likely to 320

have a diet that is in line with recommendations about both a regular eating frequency and the 321

M AN US CR IP T

AC CE PT ED

24 consumption of vegetables, fruit, and berries. Thus, from a nutritional education standpoint, 322

enjoyment of food and eating can be seen as a desirable property and should be encouraged, 323

especially in those children scoring low in this feature. This is also in line with the concept of 324

eating competence which is based on the Satter Eating Competence Model (Satter, 2007).

325

Eating competence has been related to healthier family eating patterns, such as greater meal 326

frequency and higher consumption of vegetables and fruit among 10-17 year-old adolescents 327

(Tilles-Tirkonen et al., 2015). It is therefore important that parents create a supportive eating 328

environment and promote healthier eating habits by sharing family meals avoiding restraint 329

eating attitude, and providing a positive example to their children since these traits have been 330

reported to enhance the child’s eating competence and promote healthier eating patterns 331

(Galloway et al., 2005; Gillman et al., 2000; Tilles-Tirkkonen et al., 2015).

332

On the other hand, enjoyment of food and food responsiveness have previously been linked to 333

a higher body adiposity in children (Birch & Fisher, 1998; Carnell & Wardle, 2007, 2008;

334

Fildes et al., 2015; Wardle et al., 2001; Webber et al., 2009) suggesting that these eating 335

behaviour tendencies may also be related to vulnerability to overweight, perhaps due to the 336

risk of excessive food consumption and energy intake (Carnell & Wardle, 2007). Therefore, 337

children with high scores in enjoyment of food and food responsiveness might benefit from 338

dietary counselling to avoid excessive food consumption, especially of energy-dense foods.

339

Previous studies have found that food fussiness, satiety responsiveness, and slowness in 340

eating in children are inversely associated with preferences for vegetables and fruit (Fildes et 341

al., 2015). A recent systematic review concluded that fussy eating was associated with the 342

consumption of a limited number of foods, particularly with the restricted consumption of 343

vegetables (Taylor et al., 2015). Moreover, fussy eaters have been reported to have a lower 344

intake of energy, total fat, protein (Dubois et al., 2007), and dietary fibre (Galloway et al., 345

2005), especially from vegetables (Taylor et al., 2016), and to be at a greater risk for not 346

M AN US CR IP T

AC CE PT ED

25 obtaining enough vitamins C and E from their diet (Galloway et al., 2005) in comparison to 347

non-fussy eaters. In the children examined here, we found that higher scores in food fussiness 348

and satiety responsiveness were associated with a lower consumption of vegetables. Thus, 349

children with these eating behaviour tendencies may not consume enough vegetables and 350

fruit, which could place them at a greater risk for having an inadequate intake of several 351

nutrients. There could be several reasons contributing to this behavioural trait. The inverse 352

association between food fussiness and the consumption of vegetables and fruit could be 353

explained by environmental factors related to family eating patterns, such as less availability 354

of different foods and thus, limited exposure to the taste of new foods, which may contribute 355

to children’s food fussiness. When children share a restricted family eating patterns and 356

concise taste preferences with their parents, children’s food and taste experiences may not 357

expand sufficiently. Texture resistance has also been associated with picky eating (van der 358

Horst et al., 2016) and thus, sensory aspects may play an important role in the relationship 359

between food fussiness and reduced consumption of vegetables and fruit. Moreover, the 360

results of a recent study revealed that also certain genes have an effect on the association of 361

fussy eating with reduced preference for vegetables and fruit (Fildes et al., 2016). In addition, 362

some children may have a more innate taste preference to sweet and salty and a more innate 363

aversion for bitter taste, such as vegetables, than some other children (Drewnowski, 1997).

364

To improve the acceptance of vegetables and fruit and enhance the diet quality in fussy 365

children, these foods should be repeatedly offered to children and they should be encouraged 366

to taste them (Wardle, Cooke, et al., 2003; Wardle et al., 2003). For example, providing the 367

child with an opportunity to take part in cooking family meals has been reported to influence 368

positively his or her vegetable preferences (Cunningham-Sabo & Lohse, 2013). In addition, 369

structured and shared family meals and role modelling of healthy eating have been suggested 370

to reduce food fussiness (Powell et al., 2016).

371

M AN US CR IP T

AC CE PT ED

26 Fussy eaters have been reported to consume less meat than their non-fussy counterparts 372

(Dubois et al., 2007; van der Horst et al., 2016). We found that higher scores in food fussiness 373

but also in satiety responsiveness and slowness in eating were associated with a lower 374

consumption of meat. In addition, higher scores in food fussiness and satiety responsiveness 375

were related to a lower consumption of cheese. A systematic literature review concluded that 376

a higher intake of protein was associated with increased feeling of satiety (Westerterp- 377

Plantenga et al. , 2012). Hence, children who do not prefer protein-rich foods and eat slowly 378

may achieve satiety and a sense of fullness after a lower consumption of these foods. We also 379

found that a higher score in satiety responsiveness was associated with a higher consumption 380

of candies and chocolate. This may also damage the quality of the children’s diet if they 381

compensate with sugary products for their lower consumption of meal-type foods. Several 382

studies have shown that fussy eating is related to a lower BMI and even underweight 383

(Galloway et al., 2005; Rodenburg et al., 2012; Webber et al., 2009). Based on all these 384

findings, it is important to recognize children with these eating behaviour tendencies and 385

provide them with dietary counselling on how they should improve the quality of their diet to 386

ensure an adequate nutrient intake.

387

A previous study found that a higher score in desire to drink was associated with a higher 388

consumption of sugar-sweetened beverages (Sweetman, Wardle, & Cooke, 2008). However, 389

we did not find any relationship between desire to drink and the consumption of these 390

beverages. Instead, we found that a higher score in desire to drink was associated with a 391

higher consumption of fat-containing milk but with a lower consumption of skimmed milk.

392

One explanation for these findings is that these children may not separate their feeling of 393

hunger from their feeling of thirst and therefore they drink more energy-dense beverages 394

since they feel hungry.

395

M AN US CR IP T

AC CE PT ED

27 One study detected no differences in eating behaviour subscales between preschool aged girls 396

and boys (Svensson et al., 2011), but few studies have examined sex differences in eating 397

behaviour traits in school-aged children. We found that girls had higher scores than boys in 398

food avoidance type subscales satiety responsiveness, emotional undereating, and slowness in 399

eating, but scores in the food approach subscales were similar in both sexes. In adults, 400

women have been reported to have higher cognitive dietary restraint scores than men 401

(Provencher, Drapeau, Tremblay, Despres, & Lemieux, 2003; Westenhoefer, Stunkard, &

402

Pudel, 1999). One explanation for this could be that women have higher concerns about their 403

body weight. It is possible that this food avoidant eating behaviour style of mothers is more 404

reflected in the eating behaviour of girls than boys. In addition, eating behaviour of girls may 405

be more sensitive to social situations in the family compared to boys. For example, Elfhag 406

and coworkers have reported that girls living in a single mother family have a higher 407

restrained eating style than girls living in a two-parent family, whereas such differences were 408

not found in boys (Elfhag & Rasmussen, 2008). We also found that the associations between 409

eating behaviour and the consumption of foods were more evident in girls than in boys. The 410

reasons for these findings are unknown. One possible explanation for these sex differences is 411

that the food selection in girls is more influenced by their eating behaviour, whereas other 412

factors, such as the need of energy due to a higher level of physical activity (Vaisto et al., 413

2014), may affect food selection more strongly in boys.

414

415

A strength of this study is the relatively large and representative sample of primary school 416

children. We assessed eating behaviour using CEBQ; its internal reliability and validity 417

against observation have been demonstrated to be good (Carnell & Wardle, 2007; Sleddens et 418

al., 2008; Viana, Sinde, & Saxton, 2008). We also found a good internal reliability in seven 419

of eight subscales of the questionnaire in our sample. In addition, we used 4-day food record 420

M AN US CR IP T

AC CE PT ED

28 compiled by the parents to assess food consumption; this has been stated to be an accurate 421

way to estimate food consumption in children (Buzzard, 1998; Crawford, Obarzanek, 422

Morrison, & Sabry, 1994). A clinical nutritionist guided and checked all food records 423

comprehensively with the family in order to minimize any potential recall biases. However, 424

some misreporting is still possible because quarter of the children had energy intake that 425

appears to be underestimated in relation to their energy expenditure. For example, parents 426

may have underestimated the consumption of generally recognized non-healthy foods or 427

overestimated the consumption of generally recognized healthy foods. However, in this 428

study, the findings did not change after exclusion of the misreporters from the analyses. One 429

limitation of this study is the large number of analyses that raises the concern that some of the 430

association may have been found by chance. However, many of the findings remained 431

statistically significant after multiple testing was taken into account. Finally, we cannot draw 432

any conclusions about causal relationships between eating behaviour traits and consumption 433

of foods due to the study’s cross-sectional design.

434

435

Conclusions 436

This study demonstrated that those children scoring higher in enjoyment of food are likely to 437

have a regular eating frequency and to consume a wide variety of foods, both nutrient-dense 438

foods, such as vegetables, and protein-rich foods, such as meat. In contrast, higher scores in 439

food fussiness and satiety responsiveness are related to a lower consumption of nutrient-rich 440

foods, such as vegetables, and therefore, children with these eating behaviour tendencies may 441

be at risk for having an inadequate intake of energy and various nutrients. This kind of 442

assessment of eating behaviour could be a good tool for identifying children who would 443

benefit from tailored dietary counselling to combat dietary shortcomings and to promote 444

children’s normal growth. More research is needed to assess factors that explain why certain 445

M AN US CR IP T

AC CE PT ED

29 features of eating behaviour are linked to consumption of certain foods. Future studies should 446

also address the associations of eating behaviour with dietary patterns, food variety, and 447

nutrient intake as well as factors that explain these associations.

448

449

Declarations 450

Acknowledgements 451

We are grateful to all children and their parents who participated in the PANIC Study. We are 452

also indebted to the entire research team for the skillful contribution in performing the study.

453

We gratefully acknowledge Dr. Ewen MacDonald for editing the language of the manuscript.

454

455

456

Funding 457

The PANIC Study was financially supported by grants from the Ministry of Social Affairs 458

and Health of Finland, the Ministry of Education and Culture of Finland, 459

the Finnish Innovation Fund Sitra, the Social Insurance Institution of Finland, the 460

Finnish Cultural Foundation, the Juho Vainio Foundation, the Finnish Foundation for 461

Cardiovascular Research, the Foundation for Paediatric Research, the Paavo Nurmi 462

Foundation, the Paulo Foundation, the Diabetes Research Foundation, the Yrjö Jahnsson 463

foundation, city of Kuopio, the Kuopio University Hospital (EVO-funding number 5031343), 464

and the Research Committee of the Kuopio University Hospital Catchment Area (the State 465

Research Funding). The work of Henna Jalkanen was supported by the Juho Vainio 466

Foundation, the Finnish Foundation for Cardiovascular Research, and the Olvi-Foundation.

467

468

Author’s contributions 469

M AN US CR IP T

AC CE PT ED

30 HJ conducted the statistical analyses and drafted the manuscript. VL, US, SK, TV, TAL and 470

AME participated in the data collection. VL, US, SK, TV, LK, TAL and AME contributed to 471

the critical revision of the manuscript. TAL planned and conceived the study and was 472

responsible for obtaining funding. HJ, VL, US, TV and AME helped in obtaining funding.

473

AME supervised drafting of the manuscript. All authors read and approved the final 474

manuscript.

475

476

Competing interests 477

The authors declare that they have no competing interests.

478

479

Consent for publication 480

Not applicable.

481

482

Ethics approval and consent to participate 483

The PANIC study protocol was approved by the Research Ethics Committee of the Hospital 484

District of Northern Savo (Kuopio, Finland). All the children and their parents provided their 485

written informed consent for the study. The clinical trial number of the PANIC Study is 486

NCT01803776.

487

488

M AN US CR IP T

AC CE PT ED

31 References

489

Birch, L. L., & Fisher, J. O. (1998). Development of eating behaviors among children and 490

adolescents. Pediatrics, 101(3 Pt 2), 539–549.

491

Buzzard, M. (1998). 24-h dietary recall and food record methods. In W. Willet (Ed.), 492

Nutritional Epidemiology (pp. 50–73). New York: Oxford University Press.

493

Carnell, S., Pryor, K., Mais, L. A., Warkentin, S., Benson, L., & Cheng, R. (2016). Lunch- 494

time food choices in preschoolers: Relationships between absolute and relative intakes 495

of different food categories, and appetitive characteristics and weight. Physiology &

496

Behavior. http://doi.org/S0031-9384(16)30115-9 [pii]

497

Carnell, S., & Wardle, J. (2007). Measuring behavioural susceptibility to obesity: validation 498

of the child eating behaviour questionnaire. Appetite, 48(1), 104–113.

499

http://doi.org/S0195-6663(06)00503-4 [pii]

500

Carnell, S., & Wardle, J. (2008). Appetite and adiposity in children: evidence for a behavioral 501

susceptibility theory of obesity. The American Journal of Clinical Nutrition, 88(1), 22–

502

29. http://doi.org/88/1/22 [pii]

503

Cole, T. J., Bellizzi, M. C., Flegal, K. M., & Dietz, W. H. (2000). Establishing a standard 504

definition for child overweight and obesity worldwide: international survey. BMJ 505

(Clinical Research Ed.), 320(7244), 1240–1243.

506

Cooke, L. J., Wardle, J., Gibson, E. L., Sapochnik, M., Sheiham, A., & Lawson, M. (2004).

507

Demographic, familial and trait predictors of fruit and vegetable consumption by pre- 508

school children. Public Health Nutrition, 7(2), 295–302.

509

http://doi.org/10.1079/PHN2003527 [doi]

510

Crawford, P. B., Obarzanek, E., Morrison, J., & Sabry, Z. I. (1994). Comparative advantage 511

of 3-day food records over 24-hour recall and 5-day food frequency validated by 512

M AN US CR IP T

AC CE PT ED

32 observation of 9- and 10-year-old girls. Journal of the American Dietetic Association, 513

94(6), 626–630. http://doi.org/0002-8223(94)90158-9 [pii]

514

Croker, H., Cooke, L., & Wardle, J. (2011). Appetitive behaviours of children attending 515

obesity treatment . Appetite, 57(2), 525–529. http://doi.org/10.1016/j.appet.2011.05.320 516

[doi]

517

Cunningham-Sabo, L., & Lohse, B. (2013). Cooking with Kids positively affects fourth 518

graders’ vegetable preferences and attitudes and self-efficacy for food and cooking.

519

Childhood Obesity (Print), 9(6), 549–556. http://doi.org/10.1089/chi.2013.0076 [doi]

520

Currie, C., Morgan, A., Currie, D., de Looze, M., Roberts, C., Samdal, O., … Barnekow V., 521

V. kow. (2012). Social Determinants of Health and Well-Being among Young People.

522

Health Behaviour in School-Aged Children (HBSC) Study: International Report from 523

the 2009/2010 Survey; WHO Regional Office for Europe: Copenhagen, Denmark.

524

Diethelm, K., Jankovic, N., Moreno, L. A., Huybrechts, I., De Henauw, S., De Vriendt, T., … 525

Group, H. S. (2012). Food intake of European adolescents in the light of different food- 526

based dietary guidelines: results of the HELENA (Healthy Lifestyle in Europe by 527

Nutrition in Adolescence) Study. Public Health Nutrition, 15(3), 386–398.

528

http://doi.org/10.1017/S1368980011001935 [doi]

529

Drewnowski, A. (1997). Taste preferences and food intake. Annual Review of Nutrition (Vol.

530

17). http://doi.org/10.1146/annurev.nutr.17.1.237 531

Dubois, L., Farmer, A. P., Girard, M., & Peterson, K. (2007). Preschool children’s eating 532

behaviours are related to dietary adequacy and body weight. European Journal of 533

Clinical Nutrition, 61(7), 846–855. http://doi.org/1602586 [pii]

534

Elfhag, K., & Rasmussen, F. (2008). Food consumption, eating behaviour and self-esteem 535

among single v. married and cohabiting mothers and their 12-year-old children. Public 536

M AN US CR IP T

AC CE PT ED

33 Health Nutrition, 11(9), 934–939. http://doi.org/10.1017/S1368980008002449

537

Elmadfa, I., Meyer, A., Nowak, V., Hasenegger, V., Putz, P., Verstraeten, R., … Cattaneo, A.

538

(2009). European Nutrition and Health Report 2009. Annals of Nutrition & Metabolism, 539

55 Suppl 2, 1–40. http://doi.org/10.1159/000244607 [doi]

540

Eloranta, A. M., Lindi, V., Schwab, U., Kiiskinen, S., Kalinkin, M., Lakka, H. M., & Lakka, 541

T. A. (2011). Dietary factors and their associations with socioeconomic background in 542

Finnish girls and boys 6-8 years of age: the PANIC Study. European Journal of Clinical 543

Nutrition, 65(11), 1211–1218. http://doi.org/10.1038/ejcn.2011.113 [doi]

544

Eloranta, A. M., Lindi, V., Schwab, U., Tompuri, T., Kiiskinen, S., Lakka, H. M., … Lakka, 545

T. A. (2012). Dietary factors associated with overweight and body adiposity in Finnish 546

children aged 6-8 years: the PANIC Study. International Journal of Obesity (2005), 547

36(7), 950–955. http://doi.org/10.1038/ijo.2012.89 [doi]

548

Fildes, A., Mallan, K. M., Cooke, L., van Jaarsveld, C. H., Llewellyn, C. H., Fisher, A., &

549

Daniels, L. (2015). The relationship between appetite and food preferences in British 550

and Australian children. The International Journal of Behavioral Nutrition and Physical 551

Activity, 12, 114–116. http://doi.org/10.1186/s12966-015-0275-4 [doi]

552

Fildes, A., van Jaarsveld, C. H., Cooke, L., Wardle, J., & Llewellyn, C. H. (2016). Common 553

genetic architecture underlying young children’s food fussiness and liking for vegetables 554

and fruit. The American Journal of Clinical Nutrition, 103(4), 1099–1104.

555

http://doi.org/10.3945/ajcn.115.122945 [doi]

556

Galloway, A. T., Fiorito, L., Lee, Y., & Birch, L. L. (2005). Parental pressure, dietary 557

patterns, and weight status among girls who are “picky eaters.” Journal of the American 558

Dietetic Association, 105(4), 541–548. http://doi.org/S0002822305001410 [pii]

559

Gillman, M. W., Rifas-Shiman, S. L., Frazier, A. L., Rockett, H. R., Camargo Jr, C. A., Field, 560

M AN US CR IP T

AC CE PT ED

34 A. E., … Colditz, G. A. (2000). Family dinner and diet quality among older children and 561

adolescents. Archives of Family Medicine, 9(3), 235–240.

562

Hoppu, U., Lehtisalo, J., Tapanainen, H., & Pietinen, P. (2010). Dietary habits and nutrient 563

intake of Finnish adolescents. Public Health Nutrition, 13(6A), 965–972.

564

http://doi.org/10.1017/S1368980010001175 [doi]

565

Jaaskelainen, A., Schwab, U., Kolehmainen, M., Pirkola, J., Jarvelin, M. R., & Laitinen, J.

566

(2013). Associations of meal frequency and breakfast with obesity and metabolic 567

syndrome traits in adolescents of Northern Finland Birth Cohort 1986. Nutrition, 568

Metabolism, and Cardiovascular Diseases : NMCD, 23(10), 1002–1009.

569

http://doi.org/10.1016/j.numecd.2012.07.006 [doi]

570

Kyttala, P., Erkkola, M., Kronberg-Kippila, C., Tapanainen, H., Veijola, R., Simell, O., … 571

Virtanen, S. M. (2010). Food consumption and nutrient intake in Finnish 1-6-year-old 572

children. Public Health Nutrition, 13(6A), 947–956.

573

http://doi.org/10.1017/S136898001000114X [doi]

574

Lambert, J., Agostoni, C., Elmadfa, I., Hulshof, K., Krause, E., Livingstone, B., … Samartin, 575

S. (2004). Dietary intake and nutritional status of children and adolescents in Europe.

576

The British Journal of Nutrition, 92 Suppl 2, S147-211.

577

http://doi.org/S0007114504002314 [pii]

578

McCarthy, E. K., Chaoimh, C. N., Murray, D. M., Hourihane, J. O., Kenny, L. C., & Kiely, 579

M. (2015). Eating behaviour and weight status at 2 years of age: data from the Cork 580

BASELINE Birth Cohort Study . European Journal of Clinical Nutrition.

581

http://doi.org/10.1038/ejcn.2015.130 [doi]

582

Powell, F., Farrow, C., Meyer, C., & Haycraft, E. (2016). The importance of mealtime 583

structure for reducing child food fussiness. Maternal & Child Nutrition.

584