Introduction

Obese children are at increased risk of becoming obese adults and of developing chronic diseases (type 2 diabetes and cardiovascular diseases) [1]. The preschool age is an important period for adopting healthy lifestyle behaviours including eating and physical activity habits [2]. Early adiposity rebound is strongly associated with increased body mass index (BMI) [3] and fatness in adolescence and forms a critical window of opportunity where the above-mentioned modifiable lifestyle behaviours can be targeted and potentially improved [4]. Parents and caregivers have a significant influence on the eating and physical activity habits of children. Families are such a prominent part of children’s social environment, especially at such young ages; they may play an important role in determining sedentary behaviours as well as food intake [5].

Sedentary behaviour is defined as any waking behaviour characterized by low energy expenditure (≤1.5 METs) [6]. Evidence suggests that the type of sedentary behaviour such as screen-based sedentary behaviours (e.g. TV watching, use of video games) may be more important in predicting childhood obesity to overall sedentary time [7]. A cross-sectional study in the Netherlands suggested that children (4–13 years) who watch TV > 1.5 h a day were 1.65 times more likely to be overweight [8]. There are few studies examining the relationship of multiple screen-based sedentary behaviours and the consumption of foods and beverages in young population groups [9]. In children and adolescents, TV watching is associated with lower fruit and vegetables and higher fizzy drinks consumption [10]. In European adolescents, Santaliestra-Pasias et al. [9] recently reported that increased TV viewing, computer and internet use were associated with higher odds of fizzy drinks consumption and lower odds of consuming fruit. During TV time, adolescents with high daily TV watching had high consumption of energy-dense foods and beverages such as fizzy drinks, savoury snacks and pastries [11].

Food consumption has been also shown to be determined by socioeconomic status (SES) such as parental education, household income and parental occupation [12, 13]. Children of parents from low income, occupation and education population groups tend to consume more sweets, fizzy drinks and less fruit and vegetables than their counterparts [14].

The aim of this study was to examine the association between time spent on different sedentary behaviours such as watching TV and DVDs, playing computer/video games and quiet play/activities such as looking into books, colouring, playing with dolls and/or cars and the consumption of a number of foods and beverages included in the food frequency questionnaire used in the baseline sample of the ToyBox-study. To the authors’ knowledge, this is the first study conducted in European preschool population addressing such associations.

Methods

Study design

The ToyBox-study (www.toybox-study.eu) is a cluster-randomized study aiming to prevent obesity in preschool children and their families. It was conducted in six European countries, namely Belgium, Bulgaria, Germany, Greece, Poland and Spain [15]. The detailed protocol is described elsewhere [15, 16].

In total, 309 kindergartens and 7056 children aged 3.5–5.5 years were recruited at baseline [17]. Of those, 6431 children (51.8 % males) were included in the current study. Information on maternal education was obtained from a self-administered questionnaire (study’s core questionnaire [18, 19]) including the number of years the mother studied. Mother’s education level was divided in five categories such as less than 7 years, 7–12 years of studies, 13–14 years, 15–16 years and more than 16 years of mother studied. Parents/caregivers were informed about the study, and written informed consents were obtained. The ToyBox-study adhered to the Declaration of Helsinki and the conventions of the Council of Europe on human rights and biomedicine. In all countries, ethical approval was obtained from their respective ethical committees and local authorities [17].

Anthropometric measurements

Data collection was carried out in May–June 2012. Height and weight were measured by trained personnel. Weight was measured in underwear and without shoes with an electronic scale (Type SECA 861 or SECA 813) to the nearest 0.1 kg, and height was measured barefoot in the Frankfort plane with a telescopic height instrument Type SECA 225 or SECA 214 to the nearest 0.1 cm [20]. Intra-observer technical error had a reliability above 99 %, and inter-observer technical error had a reliability higher than 98 % [20]. BMI (kg m−2) was calculated [21].

Sedentary behaviours

Data on children’s sedentary behaviour were collected via a standardized proxy-administered questionnaire (i.e. Primary Caregivers’ Questionnaire). Detailed information on the development of this questionnaire and on the data collection procedure is given elsewhere [18, 19]. Behaviours assessed included watching TV and DVDs, playing computer/video games and quiet play/activities such as looking into books, colouring, playing with dolls and/or cars [18]. Parents/caregivers reported frequency both for weekdays and weekend days. The frequency categories included: ‘never’, ‘less than 30 min/day’, ‘30 min to 1 h/day’, ‘1–2 h/day’, ‘3–4 h/day’, ‘5–6 h/day’, ‘7–8 h/day’, ‘8 h/day’ and ‘more than 8 h/day’. These answers were further aggregated into two categories including ≤1 h per day and >1 h per day. These categories are based on the Australian and Canadian sedentary behaviour recommendations for children stating that preschool children should limit their screen time to maximum 1 h per day [22, 23]. Average hours per day of TV/video viewing and personal computer use separately for weekdays and weekend days were summed up to obtain the screen time. Reliability of the sedentary questionnaire showed lower reliability [18].

Food and beverage consumption

Food and beverage consumption was assessed using a 37-item semi-quantitative food frequency questionnaire FFQ [19]. This questionnaire was based on a previously developed and validated FFQ for Flemish preschool children by Huybrechts et al. [24] and was adapted and validated for the purposes of the ToyBox-study. Low–moderate relative validity was observed which varied by food and beverage group; for some of the ‘key’ foods/drinks targeted in the ToyBox-intervention (e.g. water and soft drinks) however, the validity was good (unpublished data). In the current analysis, the 37 original food groups from the food frequency questionnaire were merged into 21 groups. Food groups were aggregated according to their nutritional content (main nutrient of protein, carbohydrates and fat).

Of those 21, 15 were chosen and entered in the current analysis based on their association with obesity development [25]: (1) water, (2) soft drinks and light drinks, (3) fresh fruit juices and packed juices, (4) sweetened milk, (5) milk and milk products, (6) dried, canned and fresh fruits, (7) raw and cooked vegetables, (8) cakes and biscuits, (9) chocolate and chocolate spreads, (10) sugar-based desserts and pastries, (11) salty snacks, (12) fish, (13) meat, poultry and processed meat, (14) pasta and rice and (15) potatoes. Food and beverage consumption was expressed by number of portions per day.

Statistical analysis

The Predictive Analytics Software (IBM SPSS Statistics for Windows) version 20 was used to analyse the data. Statistical analysis was stratified by sex because of significant differences in sedentary behaviours and food and beverage consumption patterns between males and females founded in our sample. Initially, consumption and respective sedentary behaviours were analysed by one-way analysis of covariance (ANCOVA), adjusted for centre, maternal education and BMI. Bonferroni corrections were used for post hoc multiple comparison tests.

Additionally, a generalized linear model with the inclusion of a random intercept for study centre and maternal education was used to examine the relationship between sedentary behaviours and food consumption. Maternal education and BMI were included as covariates. Values are presented as adjusted β values (estimated unstandardized regression coefficients) and 95 % confidence intervals (CI). All statistical tests and corresponding p values lower than 0.05 were considered statistically significant.

Results

Table 1 presents descriptive information about the mean and confidence interval of age, body mass index, maternal education, sedentary behaviours and country. There were significant differences regarding maternal education level and between sexes. Also, significant differences were shown between sexes with the majority of sedentary behaviours. However, there were no significant differences regarding age and body mass index between sexes nor between countries.

Table 1 Descriptive characteristics of the European pre-schoolers sample participating in the ToyBox-study (n = 6431)
Full size table

Tables 2 and 3 present the results of the ANCOVA (means and SD) for food and beverage consumption by sedentary behaviour categories, both for males and females, respectively. In general, both in males and females, and for school days and weekend days, children spending >1 h per day watching TV or total screen time had a higher consumption of fizzy drinks, juices, sweetened milk, cakes and biscuits, chocolate, sugar-based desserts and pastries, salty snacks and potatoes than those spending ≤1 h per day. Furthermore, the results also showed a lower consumption of vegetables, fruits and fish when preschoolers spent >1 h of TV or total screen time per day.

Table 2 Analysis of covariance in food by sedentary behaviour categories in 3336 European male preschoolers participating in the ToyBox-study
Full size table
Table 3 Analysis of covariance in food by sedentary behaviour categories in 3095 European female preschoolers participating in the ToyBox-study
Full size table

Results for computer time and quiet play were less consistent in both sexes. For instance, males who spent >1 h/day on quiet play during school days significantly consumed more water, fruits, vegetables, fish, potatoes and pasta and rice. Females who spent >1 h/day on quiet play during school days consumed significantly more water, fruits, vegetables, cakes and biscuits, sugar-based desserts and pastries, pasta and rice. Similar results were found for weekend days except for several food groups as fruit or sugar-based desserts in females and potatoes in males (p > 0.005).

Table 4 presents the results of the generalized linear regression models. High total screen time both in males and females, during school days and weekend days, was positively associated with high consumption of fizzy drinks, juices, sweetened milk, cakes and biscuits, chocolate, sugar-based desserts and pastries, salty snacks and potatoes. In both males and females, during school days and weekend days, negative associations were observed for consumption of fruits and vegetables.

Table 4 Analysis of general linear model of food group consumption by sedentary behaviours categories in both male and female European preschoolers participating in the ToyBox-study (n = 6431)
Full size table

Both in males and females, watching TV for >1 h/day during school days was positively associated with fizzy drinks, juices, sweetened milk, cakes and biscuits, chocolate, sugar-based desserts and pastries, salty snacks, meat and potatoes, whereas it was negatively associated with fruits, vegetables and fish. During weekend days, results were in the same direction, but in the case of females no significance was observed for chocolate and fish, and for males no significance was observed for pasta and rice.

During school days, both in males and females, spending >1 h per day of quiet play was positively associated with consumption of water, fruits, vegetables, pasta and rice. During weekend days, in both males and females, quiet play was positively associated with consumption of water and vegetables. In males, it was also positively associated with fruits, fish, pasta and rice and in females with cakes and biscuits and meat.

Discussion

To the authors' knowledge, this is the first study to examine associations between various sedentary activities and the consumption of food and beverages in European preschoolers. Due to lack of evidence in preschoolers, findings were related to those of adolescents. The main findings of this study show that TV watching and total screen time are positively associated with energy-dense foods and beverages and negatively associated with the consumption of fruit and vegetables consistent with previously reported literature in school-aged groups [26]. For instance, a Canadian study found that TV time was negatively associated with fruit and vegetable consumption and positively associated with consumption of sweets, fizzy drinks, pastries, potato chips and juices [10]. In school-aged US children, TV time was associated with lower odds of consuming fruits or vegetables and high odds of consuming candy and fizzy drinks [27]. Moreover, a study in adolescents from Massachusetts showed that TV viewing was inversely associated with the intake of fruit and vegetables [28]. Santaliestra-Pasias et al. [9] reported an association between different sedentary behaviours and consumption of energy-dense food and beverage such as soft drinks, savoury snacks and pastries. In Australian adolescents, Pearson et al. [29] observed that TV viewing was associated with energy-dense snack consumption. In both studies, ‘healthier’ dietary patterns were consistently associated with less time spent in front of a screen and more ‘unhealthy’ dietary patterns were associated with more time spent in front of a screen [30, 31].

An often observed sedentary behaviour in preschool children is quiet play. Low levels of physical activity and high levels of sedentary behaviours are common in preschoolers as they tend to spend most of time seated or moving quietly in a limited space area [32, 33]. To our knowledge, there is no study assessing the association between quiet play and food and beverage consumption in preschool children. In our study, the association between quiet play and food and beverage consumption was different to those observed for TV watching or screen time. Our results support the notion that sedentary time in children encompasses a variety of behaviours that may or may not be associated with the consumption of specific food items. Associations between specific sedentary behaviours and food and beverage consumption seem to be dependent on age (preschool children vs school-aged children or adolescents) and/or the time in which the study was performed as media technologies have gained tremendous popularity in the past decade [34, 35].

The most relevant sedentary behaviour, in terms of its association with food and beverage consumption, is TV watching [36]. One reason could be that TV viewing acts as an enhancing factor to frequently consume energy-dense advertised foods, leading to the replacement of fruit and vegetables [37, 38]. There are specific recommendations for screen time for preschool children. The Australian and Canadian recommendations for limiting screen time media time to no more than 1 h/day for preschool children were recently established [22, 23]. In order to reduce screen time, several strategies have been suggested such as removing TV sets out of children’s bedrooms, enhancing alternatives for entertainment for preschoolers such as reading, sports, creative games and promoting activities for improving their neurodevelopment [22, 39]. Our results support these specific screen time recommendations for preschool children, as we observed significant differences in energy-dense food and beverage consumption according to meeting or not meeting screen time recommendations [36].

The current study has some limitations. This is a cross-sectional study not allowing to establish causal relationships. Generalizability of the findings is limited due to the fact that there is a specific age group studied in the current study. Dietary and behavioural information was collected via parental self-reported questionnaires, which are prone to over- or under-reporting. Consequently, the quality of diet can suffer more error, but acknowledging this limitation can help to interpret it appropriately [40]. However, it was developed/adapted and validated for the purposes of the study. In addition, assessment of energy balance-related behaviours (physical activity, diet and sedentary activities) has been shown to be difficult and complex to assess in young population groups [4]. Time spent using new technologies as tablets or internet for recreational reasons was not assessed in the current study.

The main strengths of our study include the use of a large, culturally and socioeconomically diverse sample of children from six different countries across Europe. The collected information about diet and sedentary behaviours was assessed via standardized and harmonized procedures [19]. The inclusion of a homogeneous sample of preschool children is also strength. The present study includes essential information on energy balance-related behaviours for a population group that is within a critical period in lifestyle habits acquisition. Moreover, parents and caregivers must continually be reminded of their substantial influence in setting on positive habits such as physical activity and healthy diet.

Conclusions

The present findings have important implications for further studies assessing energy balance-related behaviours in European preschool children. This study provides evidence on the associations between different sedentary behaviours and food and beverage consumption in a sample of European preschoolers. Longitudinal studies are needed to confirm our results in order to confirm the associations between changes during time in the studied variables. From a public health point of view, it is important to identify screen time alternatives and supply healthy foods in order to provide an appropriate environment for obesity prevention in children. Our results support evidence calling for limiting children’s exposure to screen-based activities associated with energy-dense food consumption.