Discussion
This study has described mealtime behaviours of young Australian children, aged six months to six years, by assessing some of the less commonly identified characteristics of family meals, likely to be associated with improved diet.
Given that the evening meal is the most commonly researched family meal type [
21,
37,
43], it is important to understand its frequency, particularly as in Australia, this is the meal families traditionally share. Results from the current study reflect this, with shared evening meals the most commonly reported meal, and the meal most parents share with their child every night. This finding is consistent with the only other Australian study reporting family meal frequency within this age group [
16]. These two studies highlight that a large proportion of Australian families are frequently engaging in evening family meals. Importantly however, the current study also highlights that almost a quarter of parents reported eating with their child less than five evenings per week. This may be due to factors such as parental working hours, family size, varying ages and extra-curricular commitments of other siblings and or parents. Further research is warranted to determine what prevents families from eating meals together. Although the evidence base is smaller in younger children, regularly eating the evening meal together appears to be important for child health [
6,
15,
16]. Giving children every opportunity to be exposed to modelling at mealtimes, particularly the modelling of the consumption of vegetables may be an important way to develop and encourage healthy eating behaviours and habits. To our knowledge, this is the first study within this age group to assess family meal frequency in Australia, by differences in Socioeconomic patterning. Fewer family meals with everyone from the household present were found to be associated with the responding parent having a university education. Higher SEP is not often associated with less healthy behaviours however, this finding is consistent with some of the previous research focused on SEP and family meal frequency [
13,
44,
45]. An interplay of factors such as such as parental working hours, family size and partners’ education level is likely to influence family meal frequency and the association with SEP, but further research is warranted to better understand this.
Apart from the evening meal, it is also important to understand the frequency and location of family meals at other times of the day [
41] as these may also offer opportunities for children to establish healthy eating behaviours. Snacks as a shared mealtime may provide the best potential for nutrition promotion given that, as the current study identified, snacks are least frequently shared with parents and most commonly consumed in a less than optimal location. Changing children’s snacking behaviours, namely by increasing fruit and vegetable intakes to displace discretionary foods, is important in the context of improving child health. Research from the US suggests that changes in snacking behaviour may be a contributing factor to increasing weight trends in young children [
46‐
48] and should be an area for targeted health promotion. Although it may not be convenient for parents to eat snacks together as a family, especially for working parents, more benefits may be conferred to children if parents viewed snack time as an important time for eating together, role modelling, and opportunity for the consumption of nutritious foods.
The location of eating is also an important part of mealtime context for children [
41]. Previous studies have briefly explored family traditional dining locations, such as sitting at the table, kitchen or dining area [
16,
20,
40] yet very few studies have assessed the most common locations that children consume different meal types throughout the day. In Australia, location of other shared meals has not been assessed for this age group, until now. In the current study, children frequently ate in the car or while moving around the house. This was particularly evident for snack consumption. This indicates that the benefits of both sharing the mealtime and eating in an ‘optimal’ location are not being conferred, offering a two-fold disadvantage. Thus, promotion of nutrition through mealtime settings should focus on location as an important aspect of family mealtime occasions. Furthermore, the current study indicated that parents of higher SEP were more likely to report that their child ate at a table or bench for each meal type, consistent with the one other related study [
49]. Socioeconomic differences are likely to be multifaceted in nature and may be linked to the socioeconomic patterning of parent rated importance of family meals, as well as practical considerations such as having a room in the house allocated to dining [
20].
In addition to location, the context of eating is an important part of mealtimes for children [
41]. The frequency of children watching TV whilst eating meals is important to understand, given this has been associated with consuming more discretionary foods and fewer fruits and vegetables [
15,
50]. In the current study around one third of parents reported that their child watched TV during meals at least once per day. Similarly, in an Australian sample of four to 12 year old children, 41% of families had the TV on during every evening meal [
22]. Another Australian study reported that three to five year old children ate dinner in front of the TV an average of 2.2 evenings per week [
16]. It is important to acknowledge that families of less educated parents in the current study were more likely to watch TV during meals, highlighting the importance of tailoring family meals interventions to groups most at risk of poor diet and related behaviours. Exploring the reasons behind the socioeconomic patterning of mealtime behaviours, in particular TV viewing during meals and eating location will be an important strategy for appropriately targeting lower SEP families in promoting healthy family mealtimes.
The participant sample included more highly educated participants (71% tertiary educated) than the average Australian adult population [
51]. Although this is a common occurrence in research [
52], it means results may not be generalizable to the whole population. Recruiting online was an efficient and effective method however, more research into ways to use this recruitment method to reach more diverse groups would be valuable and would improve the generalisability of such research in future. A further recruitment challenge was that, the vast majority (97%) of participants were mothers, despite the fact that all parents were invited to participate in the study. This highlights the difficulty of recruiting fathers in research, and the importance of future research exploring the roles of fathers during family mealtimes [
53]. It is also acknowledged that online, self-report surveys can be affected by respondent interpretation of questions, social desirability bias and self-selection (non-random) bias. While efforts were made in study advertising to recruit participants with diverse views about family meals, the study may have attracted those with highest interest and motivation related to this topic. This study also required participants to have access to the internet. Though internet access in Australia is high (96% of families with children under 15 years in 2012–13 [
54]), those without internet access are unrepresented, and are most likely the lowest socioeconomic groups. Finally, this study is cross sectional and encompasses a large age range of children (pre-schoolers who have started solids). Research in this area would be strengthened by the use of longitudinal designs and studies that focus investigations within smaller age ranges.
An important strength of this study was the development of reliable, purpose designed questions. The test-retest analysis of these questions showed moderate to good/substantial reliability [
42] and as such provide reliable survey measures. Further, the online recruitment and survey design of this study enabled rapid, low cost data collection with all advertisement voluntarily (no cost) displayed on popular parenting online sites. Recruitment occurred over a period of less than three months.