1 INTRODUCTION

Children's packed lunch boxes have a disproportionate amount of discretionary foods.^{1, 2} The 2011–2012 National Nutrition and Physical Activity Survey showed children consumed 37% of their total daily energy intake during school hours, and 44% was from discretionary foods.² Of the discretionary foods consumed during school hours, the most frequently consumed foods were sweet biscuits, processed meats, savoury biscuits and muesli or cereal bars.²

Parents are concerned about what to put in lunch boxes but are influenced by factors including convenience, child preference, cost and food safety.^3-5 Packaged foods, particularly single serve multipacks, meet some of those concerns. However, the food package is a source of promotion for the product at the point of purchase and can attract the attention of parents and children and influence purchase decisions.^{6, 7} Children are particularly vulnerable to food marketing and promotion influences, and it has been shown to affect children's food preferences and intake.⁸

Chile has implemented regulation that limits the use of marketing techniques that appeal to children such as cartoons, mascots and incentives on food that is high in energy, sodium, saturated fat and total sugar, including on packaging.⁹ Products high in any of these nutrients also carry a “high in” warning label.⁹ In Australia, government regulation of food marketing to children is limited to children's television programs. Otherwise, there are advertising industry codes that specifically mention advertising to children, but these codes are not aligned with best practice recommendations and have limited impact on reducing children's exposure to marketing of unhealthy foods.¹⁰ If Australia were to develop regulation similar to Chile, there are several options on how to define foods not suitable to carry child-directed marketing. Australian governments, through the Health Council, have developed and endorsed a food category-based guide to accompany food marketing guidelines, but it has not yet been implemented in any settings.¹¹ World Health Organization (WHO) regions, including the Western Pacific, have also developed criteria for assessing food suitable for marketing to children.¹² Many products already carry a Health Star Rating (HSR), an Australian government-developed voluntary front-of-pack scheme that categorises the “healthiness” of packaged foods from half a star (least healthy) to five stars (most healthy).¹³

Given the high proportion of discretionary foods in Australian children's diets, this study aimed to (a) investigate the use of on-pack child-directed and parent-directed marketing techniques on lunch box snacks and (b) use several nutrition criteria to determine the “healthiness” of those snacks. These results can inform policy recommendations.

2 METHODS

Packaged snacks directed at parents as appropriate school lunch box snacks were identified on “pantry,” “kids snacking” and “lunch box” sections of the three major Australian supermarket websites (Coles, Woolworths and Aldi). Data collection was independently undertaken by two researchers and entered into an Excel spreadsheet for analysis. Data were cross-checked by each of the two researchers and another researcher independently checked all entries and any anomalies were discussed and resolved. Information extracted included serve size, pack size, nutrition information (energy, saturated fat, total sugars, sodium, dietary fibre and protein), percentage of fruit, vegetables, nut and legumes (FVNL) and the presence of non-nutritive sweeteners. Where fibre was not described on the nutrition information panel (NIP), it was estimated using ingredient composition and the Australian Food, Supplement and Nutrient Database 2011–2013. Where FVNL was not described, it was estimated using the ingredient list and comparison with like products.

The use of child-directed and parent-directed (including nutrition claims) marketing tactics on the pack was also recorded. The number of products that carried at least one marketing tactic was recorded. Criteria for child-directed was based on previous Canadian research¹⁴ and included any of the following: child-directed product name, shape, flavour or colour, the presence of child-directed font, images or language on the product packaging. Parent-directed marketing included any claims that encouraged parents to purchase the product for their children or to include it in lunch boxes (e.g., lunch box friendly and images of lunch boxes) and any nutrition-related marketing claims including reference to the presence, absence or amount of any nutrient or ingredient (e.g., high in fibre and no artificial colours).

The healthiness of each product was assessed using different criteria: the HSR¹³ (using <3.5 as the cut off for “unhealthy”),¹⁵ the Chilean criteria,⁹ the Australian Health Council Guide¹¹ and WHO Western Pacific nutrient profile model.¹² A HSR <3.5 is a commonly used threshold,^{15, 16} including in the New South Wales government-developed Healthy School Canteen Strategy.¹⁷ Snacks were classified into savoury or sweet snacks then subcategories (Table 1).

Descriptive statistics were calculated in the Microsoft Excel.

3 RESULTS

The sample contained 79 savoury snacks and 56 sweet snacks (Table 1). The average HSR of the 135 products was 2.2 with 79% of products having a HSR <3.5. About 79% of products had a parent-directed marketing feature and an average HSR of 2.3. The average HSR for the 39% of products with child-directed marketing was 2.2. In all but one of the subcategories, the average HSR for products with parent-directed marketing was equal to or higher than the category average. For products carrying child-directed marketing, five of the eight subcategories had a lower HSR than the subcategory average (Table 1). The subcategories of savoury biscuits, sweet biscuits, sweet dairy, confectionary and fruit all had a lower average HSR if they carried child-directed marketing than the category average.

The most common child-directed marketing technique was the use of cartoon characters (21% of all products). For example, the packaging of one snack containing cheese, mini cookies and processed meat featured licenced characters from a popular animated television series and G-rated movie. A not-for-profit body representing children's interests in media assessed this animated series as suitable for children of all ages, with its main audience being under the age of seven.¹⁸ Parent-directed marketing included claims such as “no artificial colours/preservatives/flavours” (20%), “no added sugar” (3%) and phrases linked to providing food for children, such as “lunch box friendly” (4%), “fun” (4%) and “minis” (4%).

Marketing would be removed from 89% of sweet snacks and 91% of savoury snacks if products with a HSR <3.5 were not allowed to carry child-directed marketing (Table 2). The Health Council Guide was slightly stricter; using it as the criteria for carrying child-directed marketing would remove that marketing from 93% and 92% of sweet snacks and savoury snacks, respectively. Child-directed marketing would be removed from all products that now carry that marketing if the Chilean criteria was used and most products if the WHO Western Pacific Region nutrient profile model was used. For products with parent-directed marketing, between 77% and 86% of products would not be allowed to use that marketing if either the HSR or the Health Council guide was used. A small percentage of products carrying parent-directed marketing would be allowed based on the Chilean or WHO criteria (Table 2).

4 DISCUSSION

Snacks marketed as ideal for children's lunch boxes were mostly unhealthy foods (79%). We found that products with child-directed marketing techniques represented almost 40% of the sample and, in five of the eight food categories, these products had a lower HSR than the overall sample. Not unexpectedly, almost 80% of products included marketing targeting parents with claims including “lunch box friendly.” The products carrying parent-directed marketing had a similar HSR to the category average. These results indicate that child-directed marketing was more commonly used on less healthy options.

We found that 90% of snack products could not carry child-directed marketing if a HSR threshold of <3.5 was used in a policy to protect children from such marketing, and 99% if using the WHO Western Pacific region criteria. While sampling and profiling are different across studies, studies around the world have found similar results. An Australian study looked at products from five food manufacturers and one retailer and found 45% had HSR <3.5 and 47% of all packages were designed to appeal to children.¹⁶ A United Kingdom study of products specifically marketed to children found over 90% used cartoon characters and overall 41% of the products were unhealthy based on a United Kingdom nutrient profile model.¹⁹ A Canadian study, using the WHO Regional Office for Europe criteria, found 88% of child-directed products were unhealthy, and nutrition content claims and cartoon characters were commonly used on packaging.¹⁴ A study of the Slovenian food supply also used the WHO Regional Office for Europe Model and found 93% of all products with child-focused marketing and 73% of products without were unhealthy.²⁰

Packaging plays a role in influencing purchases and marketing strategies have an influence on children's food requests and parent's choices.^{5, 6} Intercept interviews with parents leaving a supermarket with a child found that 70% had purchased at least one item requested by their child, with 88% of requested foods being unhealthy.²¹ An experimental Australian study investigating the impact of front-of-pack marketing on parents’ choice of cereal for their children found that written claims, particularly those promoting the product's nutritional value, significantly contributed to parents’ choice.²²

Nutrient criteria are required if governments are to implement a policy to restrict marketing techniques on unhealthy snacks that may influence children and parents' lunch box choices. We used the HSR as a way of determining “healthiness” as it is in use on packaging, albeit voluntarily. The Health Council Guide for foods not appropriate to be advertised to children offers another alternative and showed similar results to the HSR in this study. The government-developed Health Council Guide has already been endorsed by governments for use in regulating marketing to children, aligns with the Australian Dietary Guidelines and is simpler to use, being based on food categories and not requiring access to the NIP—or other details not required on the NIP such as the fibre and FVNL content.¹⁵ Additionally, it can be applied to all food products, while the HSR is on 41% of the eligible food supply.²³ The WHO Western Pacific region criteria is a hybrid criteria excluding whole categories from advertising in some cases but in others, setting limits on nutrients such as sugar, saturated fat and energy. It is more difficult to use as it requires access to the NIP for many food categories.

Another option is a system like the Chilean criteria—a front-of-pack labelling scheme indicating “high in” products with thresholds for energy, saturated fat, sugars and salt—that is used in comprehensive regulation. In Chile, it is used to restrict marketing to children in different media as well as restricting the use of child-directed marketing on packaging. The results from Chile show a significant decrease in child-directed marketing on “high in” children's cereals since the implementation of their policy.²⁴

This study is limited by the size of the sample; however, unlike some studies that solely investigated products carrying marketing, our sample covered a variety of snacks, including international snacks commonly marketed through the major food retailers for their convenience for lunch boxes. Although sliced vegetables are recommended as lunch box snacks, we have limited our sample to prepackaged snacks that require no preparation. Our sample may not be fully representative of typical lunch box contents, as we lack data on those specific food choices. The Fibre and FVNL content had to be estimated in some cases as it is not mandatory to list these except in circumstances where a related claim exists on the label; however, the method used has been used extensively in public health research.

5 CONCLUSION

While fresh fruit, vegetables and wholegrain products are the recommended snacks for children, parents have indicated convenience is a key factor in their choice of lunch box snacks.⁵ Removing child-directed marketing from unhealthy products would decrease “pester power” experienced by parents when shopping and contribute to addressing the high proportion of discretionary foods eaten by children at school.² This initiative, alongside a change to government policy to mandate the HSR, would provide more helpful guidance on healthier snacks. In a focus group study, parents said that the government had a role in developing labelling requirements to show the healthiness of products.⁵ These two policies would complement school- and community-based interventions such as app-based lunch box interventions²⁵ and shelf tags that highlight the healthiest packaged foods in a supermarket²⁶ to help parents replace discretionary foods with foods aligned with dietary recommendations.

While our study focused on lunch box options, regulation should comprehensively cover child-directed marketing on packages and not only child-targeted products.⁶ As well, the regulation of child-directed marketing on packaging should form part of comprehensive regulation to protect children from the marketing of unhealthy foods.²⁷

AUTHOR CONTRIBUTIONS

Wendy L Watson conceptualised the study design. Sophia Torkel, Martha Kat and Wendy L Watson carried out data collection. Wendy L Watson, Sophia Torkel and Martha Kat conducted the data analysis. All authors contributed to the interpretation of results. Wendy L Watson drafted the initial manuscript. All authors critically reviewed the manuscript and approved the final version submitted for publication.

FUNDING INFORMATION

This research was funded by Cancer Council NSW and received no other funding.

CONFLICT OF INTEREST STATEMENT

The authors have no conflicts of interest to declare.