Treats: low socioeconomic status Australian parents’ provision of extra foods for their overweight or obese children

Key words: extra foods, food provision.

In Australia, 25% of children are overweight or obese.¹ Within lower socioeconomic status (SES) groups in Australia, the problem is more pronounced, with 32% of children classified as either overweight or obese.² Dietary guidelines exist to assist populations consume appropriate amounts and types of food, reflecting the importance of diet as a contributor to weight status.³ In Australia, a new set of guidelines has been released.⁴ The guidelines provide serving size recommendations for the five food groups. Another category of foods, ‘extra foods’, is mentioned, which relates to foods that are both energy-dense and nutrient-poor (EDNP).

Overconsumption of extra foods is more common among low SES children compared with their higher SES peers,^5,6 which may be contributing to the higher rates of overweight and obesity observed in lower SES children.^2,7 This raises the question of why low-SES parents may be treating their children comparatively more frequently with extra foods.⁸ The objective of the present study was therefore to explore low SES parents’ beliefs and behaviours relating to the provision of treat foods (i.e. extra foods) for their overweight children.

Previous research investigating personal and environmental influences on parents’ food provision behaviours has identified several relevant factors that have implications for treating behaviours. For example, Petrunoff and colleagues identified the influence of the child, parenting practices, health considerations, convenience and cost, child care influences, parental perception of influences on their children, and social influences, as influencing parental treat provision, all of which were relevant to both low and high SES parents.⁸ Factors affecting the food provision behaviours of low SES groups in general include food insecurity, unemployment, low income, low educational attainment, nutrition knowledge, health consciousness, lack of time, and the perceived cost of healthy eating.^9–12 As low SES families are more likely to have children that are overweight and obese,² it can be difficult to distinguish food provision behaviours that are the result of socioeconomic circumstances from those that are a response to a child’s weight status. Hughes and colleagues¹³ reported that low-income mothers with overweight or obese children fed their children extra foods, despite their weight status. This was often in response to beliefs that food should be used as a mechanism for behavioural control, that food is a necessary comfort in stressful situations, and that genetics largely determine children’s weights.

Previous research has shown that there is considerable scope for improving children’s eating behaviours.^14–18 In Western Australia, the context of the present study, 34% of boys and 35% of girls consumed snacks or meals from fast food chains at least once per week. Confectionary and cereal bars were consumed daily by 33% of boys and 39% of girls. Cereal-based products (such as biscuits, cakes and pastries) were consumed daily by 71% of boys and 77% of girls. Snacks such as potato chips, corn chips and pretzels were consumed daily by 27% of boys and 28% of girls.¹⁶

Petrunoff et al.⁸ explored parents’ understanding of extra foods and found that these foods were frequently provided to their children. Despite their frequent provision, they were referred to with names suggesting that they should be provided only occasionally (i.e. ‘sometimes foods’ and ‘treats’). Treat foods have been described as unhealthy foods,¹⁹ foods that can control children’s behaviour,²⁰ sometimes foods, junk,⁸ novel foods and unnecessary foods.²¹

Examples of treats, as identified by parents, have included ice cream, pizza, chips, and foods from fast-food restaurants.^8,22 Parents have been found to believe that treat foods are primarily unhealthy foods that are well liked by children, and that such foods should only be consumed occasionally.^22–24 In other studies, frequent treat provision has been reportedly considered acceptable.^8,19 In Hesketh et al.’s study,¹⁹ parents believed that children should be allowed to have treats; however, the frequency with which they believed treats should be consumed varied. Most parents felt that children should be allowed daily treats, whereas a minority felt that one per week was sufficient. In another study, parents were found to hold the belief that treats could be provided on a frequent basis if children’s diets were balanced with healthy foods.⁸ In addition, Hill²⁰ explained that treat foods are used to control behaviour and can be given as a sign of affection or withdrawn as a form of punishment. This study focuses on understanding why low SES families may treat their overweight or obese children. Although treating is common across all socioeconomic groups, there may be issues of particular importance to low SES groups that could have implications for addressing children’s weight problems.

Parents from Perth were recruited using a social research agency as part of a larger study on child health. Those eligible to take part met the following criteria: having an overweight or obese child aged between 5 and 9 years (as per Cole et al.’s²⁵ body mass index cut-offs), no education past high school, and a gross annual household income under AU$60 000 (the average household income in Australia is AU$76 000).²⁶ Of the sample, two-thirds were single parents and around one-third were parents living on government pensions; many of the participants had very low incomes. Table 1 provides an overview of the sample.

Table 1.  Demographic profile of study participants

In total, 37 participants enrolled in the study, of which 35 were women and two were men. At the end of the 12-month data collection period, 17 parents remained actively involved. Ethics clearance was obtained from a University Human Research Ethics Committee. All participants gave their written consent to take part in the project.

Information was collected from participants via the use of self-introspections, interviews, and focus groups. Participants were expected to take part in regular interviews and one focus group, and to provide fortnightly introspections. The interviews were carried out in the first, fourth and final months of the study. The focus groups were conducted six months into the study to facilitate member-checking of the findings gathered up until that point. Data collection was flexible to accommodate all participants. Parents could submit different numbers of introspections and could take part in variable numbers of interviews and focus groups, depending on their preferences: for example, small focus groups were conducted on a monthly basis for three participants who lived near each other and who preferred this option to providing fortnightly introspections. These participants reported being inadequately motivated to write fortnightly introspections, and preferring the structure of having someone else arrange the interview schedule and come to them to ensure they provided their contributions.

In their self-introspections, participants were asked to monitor their behaviours, thoughts and feelings related to their children’s health, particularly those relating to diet. Introspections were submitted by email, a weblog, a telephone message service, or mailed letters. The interviews and focus groups were semistructured, allowing parents to raise topics they considered to be important influences on their children’s health. The interview protocol was pilot tested with a sample of parents from the enrolled cohort, with the intention of assessing the viability of the planned requirements of the study and making any necessary changes before officially beginning. In line with feedback received, rather than interviews being conducted at the university, as originally planned, they took place in locations of the participants’ choice, which in most instances were their homes and occasionally their workplaces. Parents were remunerated with up to A$75 per month for their participation in the study.

Over the 12 months, 354 self-introspections were submitted and 72 interviews and 12 focus groups were conducted. A list of possible introspection discussion topics was provided to participants to assist them in the data provision process. Table 2 provides a broad overview of the discussion topics and prompts used for the introspections, interviews, and focus groups. Most parents used the prompts, with several adding their own topics of discussion.

Table 2.  Discussion topics and prompts used across data collection episodes by participants

The topic of treats emerged spontaneously for some parents. Participants were also asked questions in their interviews and given prompts in their introspections, such as, ‘What springs to mind when you hear the word treats?’, ‘What would you classify as a treat?’, and ‘How often do you think children should have treats?’ The transcripts were imported into NVivo9 for coding and analysis. The coding framework was developed using theoretical concepts from the literature, items listed in the interview guides, and topics emerging spontaneously from the data. A treats node was created and broken down into several subnodes (e.g. rewards, special occasions). Interpretation of findings in relation to treats was facilitated by reading full transcripts, interrogating the ‘treats’ node, and using matrix and text searches. Many nodes and transcripts were re-read several times to assist with interpretation and understanding of the topic under investigation. The matrix and text searches enabled the researchers to explore relationships between topics and to identify themes. There did not appear to be any differences in discussions or information provided on treats with respect to the data collection process (introspections, interviews, or focus groups). All names used are pseudonyms.

In this section, the results and discussion are presented together. Parents’ definitions of what constitutes a treat are discussed, followed by their reports of when and why treats are given. Table 3 provides an overview of the identified themes with supporting quotes.

Table 3.  Summary of identified themes relating to treat provision, with supporting quotes

The most common definition of a treat in the present study was an unhealthy food that should be consumed infrequently. Despite this common definition, most of the study participants reported providing treats to their children daily. Given the overweight status of the children involved in the study, the frequency with which treating occasions occurred was problematic. There was some parental acknowledgement of this. These parents reported not wanting their children to have treats, while at the same time expressing concern that their children were consuming too many treats due to social pressures surrounding treat giving. They were not able to identify a solution to this problem. The contradiction between reported behaviours and the accepted definition of treats appeared to be a reflection of the parents’ desire to control their children’s behaviour, to demonstrate love and affection, and to address deprivation beliefs.

Parents often reported using treats to control their children’s behaviour. This was done through the use of rewards and bribery. Parents most commonly reported giving their children treat foods if they agreed to eat healthy food served to them, cleaned their rooms or behaved themselves. The use of food as a behavioural control mechanism illustrates a food hierarchy within which treats appear to reside at the top for children. This finding is not exclusive to low SES parents, with similar findings reported by Petrunoff et al.⁸ in both low and high SES families, however Hendy and Williams²⁷ found that low SES mothers used more rewards than their high SES peers. The use of treats as rewards, bribery and as a sign of affection can be problematic because these behaviours teach children to overeat and to use food as an emotional crutch,^28,29 which may have implications for the children’s weights. Teaching parents about this might be beneficial. It is important, however, to understand why parents do this. Pescud and Pettigrew³⁰ provide a discussion of some of the barriers faced by this sample of parents in relation to their food provision practices (including time scarcity, cost, fear of their children experiencing hunger, and beliefs about balancing unhealthy food with healthy food), which may have implications for treating behaviours.

Emotions played an important role in the provision of treats in the present study. Providing children with treats was often considered a tangible expression of love and affection that could make both parents and children feel good. Many of the parents in the present study were living without a partner, and all were on low incomes; therefore, food was an inexpensive way to physically demonstrate love and to feel love in return. A similar finding has been reported elsewhere,³¹ with the gifting of unhealthy foods being more prevalent in lower SES families and used as a way of compensating for loss of time with a parent or deprivation of material possessions.

In this study, there was a strong desire in some parents to ensure their children did not experience deprivation in relation to the provision of treats, a finding common in the literature and across socioeconomic groups (e.g. Pagnini et al.³²). In the present study, this was especially important in terms of comparison with peer groups, as these parents did not want their children to feel as though they were missing out. Parents largely attributed their urge to treat their children to the media’s influence on children through heavy advertisement of treat foods. It is also possible that, due to their socioeconomic circumstances, this study’s parents may not have had other avenues for treating their children with more expensive items, thus more emphasis may have been placed on the use of food as a treat, as per Roberts and Pettigrew.³¹ The influence of society and the expectation to provide children with treat foods was also common in Petrunoff et al.’s study,⁸ and was expressed by parents from both low and high SES groups. The myriad food advertisements promoting the use of sweet treats to signify parental love³³ might be negatively impinging on parents’ treating behaviours and their children’s subsequent weight status. In such advertisements, parents are encouraged to treat their children, while society warns parents against letting their children put on excess weight.³³ This contradiction is problematic and needs addressing, especially in light of the high rates of child obesity.³⁴

The types of foods used as treats and the frequency with which they are given suggests a potential need for clearer information on packaged treat-type foods, which may enable parents to better decipher the nutritional status of foods purchased and whether or not they are suitable for frequent consumption. In particular, front-of-package labelling could assist parents in becoming more aware of how often their children should consume particular products. Many parents in the present study reported experiencing difficulty interpreting labels, which made it difficult for them to make informed choices when purchasing products, and this may have had implications for their children’s weights. Given the difficulties reported by many of the participants when attempting to use the mandated nutrition information on labels, they tended to rely on the potentially misleading marketing information on the front of the packages to assess a product’s healthiness. Other studies have reported that parents would welcome a traffic light labelling system to enable them to easily interpret the nutritional value of the foods they are considering purchasing.^35,36 This suggests a need for easy-to-understand food labelling and more community education surrounding the provision of extra foods.

The use of non-food treats has been reported previously;³⁷ however, the present study sheds some light on how this type of treating works in low SES families. Some parents in the present study also reported using non-food treats on occasion. For a few who were particularly time-poor, certain leisure activities (e.g. visits to the park) that might be considered everyday events to others were considered treats because of their infrequent occurrence. Health promotion initiatives could capitalise on children’s desire to be ‘treated’ in healthy ways. Emphasis could also be placed on other inexpensive non-food treats such as toys or stickers, as previous research has shown these methods to be effective as ways of treating children;³⁷ this may also have benefits for controlling children’s weights. For example, a US study found that when given a choice between candy and toys on Halloween, children were equally likely to choose either option.³⁷ This has implications for other special occasions that are traditionally celebrated with treat foods. For low SES families especially, promoting the use of inexpensive non-food treats could be particularly appealing. Given some parents’ use of specific fruits as treats, further research could be conducted to better understand how healthy foods can be used as treats. This could provide alternatives for parents who feel compelled to treat their children but are concerned about the health implications of frequently providing extra foods. It is acknowledged, however, that due to the normalisation of daily treats in most children’s diets, the use of non-food treats as a substitute would likely have limited applicability, at least in the short term. In the medium term, however, changing treating expectations and the way in which treats are positioned could offer a positive step forward for improving children’s diets and weights.

Efforts to improve children’s diets should involve population-level approaches aimed at raising awareness of dietary needs.¹⁴ Policies that positively influence children’s diets, such as the Healthy Food and Drink Policy implemented in West Australian state schools, are also valuable because they are widely accepted and supported by parents and have the potential to influence school health curricula.³⁸ Increased knowledge surrounding dietary guidelines is linked to better food choices and eating patterns.³⁹ More promotion of what is an extra food and how excessive consumption of extra foods may lead to weight gain might be warranted.⁸ This could be especially important for low SES parents who are likely to have lower levels of health literacy and nutrition knowledge.⁴⁰ Given the issues relating to why treats are given to children, more education on effective parenting styles might be warranted through programs such as the Triple P Positive Parenting Program.⁴¹

The qualitative nature of this study means that the findings cannot be generalised to the wider population as they are representative of a small sample of low SES parents (mainly mothers) with overweight or obese children. Future research could explore a larger and more socioeconomically and ethnically diverse sample, whose beliefs and behaviours might be different to those described in the present study. In addition, further work could be conducted with fathers to ascertain their experiences in feeding overweight children. Self-selection may have placed limitations on the study. For example, the monetary reimbursement may have added bias to the sample because only those who were motivated by financial incentives might have agreed to participate. However, it may not have been enough to have attracted others, particularly given the relatively long duration of the study. Those interested in health issues may have been more likely to participate. A further limitation relates to ‘social desirability bias’,⁴² whereby participants may not have presented themselves in an accurate way. In doing so, they may have given more positive accounts of their thoughts and behaviours than was always accurate.⁴³

A typology of low SES parents of overweight or obese children’s beliefs and behaviours relating to the provision of treats has been presented. Most parents appeared to understand that extra foods should be eaten infrequently; however, their behaviours often seemed contradictory to their beliefs. There is scope for improving these parents’ knowledge in relation to the provision of treat foods, especially with regard to children’s dietary and weight outcomes, while being cognisant of the barriers they experience when feeding their children.