Products that resemble foods and appeal to children Potential risks of accidental ingestion

3. What makes products appeal to children or resemble food?

The SCCS opinion states:

8. CHARACTERISTICS OF CPRF AND CAP

8.1. Characteristics of consumer products resembling food (CPRF)

As defined in the introduction, CPRF are a sub-set of consumer products, such as cosmetics and liquid household products, which in design, shape, or presentation resemble food and could mistakenly be consumed by children or the elderly.

More specifically, products that appear to be other than they are and endanger the health and safety of consumers can be defined as in the Council directive 87 /357 /EEC (Article 1, number 2): Products “which, although not foodstuffs, possess a form, odour, colour, appearance, packaging, volume or size, so that is likely that consumers, especially children, will confuse them with foodstuffs and in consequence place them in their mouths, or suck or ingest them, which might be dangerous and cause, for example, suffocation, poisoning, or the perforation or obstruction of the digestive tract.”

The scope of this opinion is limited to cosmetics and liquid household products. The present section aims to give examples of the characteristics that make cosmetics and liquid household products more food-resembling.

8.1.1. Colour

For the category of cosmetics and liquid household products, the characteristic of a food-resembling colour can be related to the packaging, or when the packing is transparent or missing, to the product itself.

Liquid foods can have very different colours, such as orange (orange juice, soft drinks), white (milk), black (cola), brown (coffee, cocoa), red and yellow (several fruit juices and soft drinks). The same is true for solid foods. It should also be noted that colours such as blue or green, which were previously reserved for non-food products such as cleaners, are now also used in foods. Due to new trends in food marketing, the frontier between food products and cosmetics has been blurred.

8.1.2. Shape, packaging, imagery

For cosmetics, the characteristic of a food-resembling shape can be related to either the product shape itself (e.g. soaps that are shaped like lemons) or to the product packaging. Product packages that resemble, in their shape, real-life containers of solid foods, such as cans, bowls, plates etc. have a food-resembling shape.

For liquid household products, the characteristic of a food-resembling shape is mainly related to the product packaging. Product packages that resemble, in their shape, real-life containers of liquid foods, such as bottles, cans, cups, glasses etc. have a food-resembling shape.

Not only the product itself or the shape of the product package can resemble a food, but also the imagery used on the packaging can create an association with food. For instance, oranges can be pictured on an orange-coloured shower gel, or lemons on a household cleaner.

Other aspects of food packaging that are displayed on a non-food product, such as fake nutrition tables, can also increase the extent to which a product is food- resembling.

8.1.3. Taste and odour

• For cosmetics and liquid household products, characteristics of odour or flavour are conveyed primarily either by imagery (see above), or by names and other written descriptions on the product. Product packages can also be opened to sample the odour directly.
• Odours, flavour and their descriptions can make a product more imitating when they closely resemble real food odours and flavours (e.g. fruity shower gels, honey lip balms) or when the description suggests that they do (e.g. “Sweet lime body butter”).
• Bitter tastes are generally not preferred, which is why bittering agents have been used to deter ingestions and poisonings. The impact of bittering agents in poisoning prevention is discussed further in section 8.3 and Annex IV.

8.1.4. Accessibility and storage

• Placement at point of sale: products are more food-resembling when they are placed close to food products.
• Storage: food-resembling products could be consumed in error when they are stored close to food. Since the elderly often live in small spaces, this is can be a contributing factor.

The characteristics of CPRF discussed above are based on descriptions of the properties of food-resembling products that could lead to poisoning. There are no studies, for any of the characteristics mentioned, that tested experimentally the likelihood of poisoning or ingestion with regard to cosmetics and liquid household products. Although there are case reports (e.g. the ingestions of a colourful cleaning product, see Annex II), there are no experimental data available that show causally that, for instance, liquid household products with an orange colour (relatively more food-resembling) are ingested more often than the same products with a blue colour (relatively less food-resembling). Nevertheless, the above mentioned characteristics can serve as proxies to evaluate whether products are more or less food-resembling, until more specific data are available.

8.2. Characteristics of child-appealing products (CAP)

CAP can also be defined as a sub-set of normal consumer products that are appealing to children by design or presentation and may therefore be consumed by children by mistake. There is an overlap between CPRF and CAP (e.g. some food-resembling products may be particularly child-appealing), but the two categories are not identical.

The scope of this opinion is limited to cosmetics and liquid household products. The present section aims to give an overview of the characteristics that make cosmetics and liquid household products appealing to children. It should be noted, however, that the appeal of a product for children cannot be defined objectively, but only in relative terms (this is different to CPRF, where it is possible to describe the extent to which a product imitates a food by comparing it to that food). Children can be attracted to nearly anything within their reach, depending on the number and type of other attractors in their environment, their situational and dispositional inclination to explore, and many other factors. For the assessment that CAP can pose a serious risk to health and safety of children, the personal and environmental risk factors presented in section 7 will therefore be of particular importance.

Examples of characteristics of child-appealing products are given below:

8.2.1. Colour

• Attractively coloured packaging may serve to influence children’s selection or persuasion in stores, and colour is also an important determinant of food liking and judgements of sweetness and other tastes (Hutchings 2003, Lavin and Lawless 1998, Léon et al. 1999).
• However, studies on colour preferences in children for different products do not show any consistent results. Examples include:
  
  • In an older study by Schneider (1977) with children between 3 and 5 years of age and empty product packages without any specific product relation, white containers led to the largest proportion of high attraction (48%) followed by black containers (33%), and finally, red containers (26%). Only these three colours were offered to the children.
  • In a study of colour preferences for different types of candies in children aged between 5 and 9 years, children preferred candies that were red, green, orange and yellow, in that order (Walsh et al. 1990). Only these four colours were offered.
  • Another study investigated colour preferences for three types of products (cereals, biscuits and drinks) with children aged 3 to 5 years. The colours chosen most frequently were pink (40.9%), purple (15%) and yellow (15%), and these colours were also among the favourite colours of the children in general. Nine different colours were offered in this study (Marshall et al. 2006).
• The results seem to be highly dependent on the type of product, choice set of colours and age of children. Children up to 5 years do not seem to be able to give repeatable results when asked for their favourite product colours (Léon et al. 1999).

8.2.2. Shape, packaging, imagery

• Products that are marketed for children generally use lots of vivid imagery, often in cartoon or comic style. Children are attracted to products that picture a cartoon character or other characters or objects that they are familiar with from other contexts, e.g. from TV or books (Ülger 2009). For instance, in a study with 4 to 6 year olds, children significantly preferred the taste of foods that had popular cartoon characters on the packaging, compared with the same foods without characters (Roberto et al. 2010). Food products presented in this study were graham crackers, gummy fruit snacks and carrots.
• Child-proof caps make it difficult for children less than 5 years of age to consume significant quantities of household chemical consumer products (Bateman 2003; see also WHO 2008a).
• In contrast to CPRF, there are no general characteristics of the shape or consistency of cosmetics and liquid household products that make them relatively more child- appealing.
• For young children, the presence of product labels or warnings will not have an effect because they cannot read or interpret them (WHO 2008a). Even though older children can read, the information that a product may not be suitable for a certain age group is not very likely to have an effect. The results of the study of Schneider (1977) even suggested that in children between 3 and 5 years of age the labelling of e.g. poison (skull and cross bones) may itself be attractive.

8.2.3. Taste and odour

• Children initially prefer sweet tastes and reject sour and bitter tastes; these are genetic predispositions (Berk 2009, Birch 1999, Birch and Fisher 1998b, Schwartz et al. 2009). Later on, their preferences for the majority of foods are shaped by repeated experiences (Berk 2009, Birch 1998a). A developmental study with 1,291 children aged from 4 to 16 years showed that across age and gender, children rated sugary and fatty foods most highly, although ratings for fruit were also high (Cooke and Wardle 2005). In this study, girls liked fruit and vegetables more than boys did; boys liked fatty and sugary foods, meat, processed meat products and eggs more than girls.
• Analyses of the type of foods marketed to children show correspondingly that these are predominantly high in sugar and fat (Elliott 2008, Story and French 2004).
• With growing age and perceptual-attentional skill, children seem to focus more on flavour (rather than colour) when asked to identify drinks (Oram et al. 1995; see also Liem et al. 2004).
• Bitter tastes are generally not preferred, which is why bittering agents have been used to deter ingestions and poisonings. The impact of bittering agents in poisoning prevention is discussed further in section 8.3 and Annex IV.
• Odour is an important cue for taste, so it can be expected that children will prefer sweet, fruity and candy-like odours.
• The study of Schneider (1977) showed that odour can also be an attractor in itself: packages with no fragrance, pleasant fragrance and antiseptic fragrance resulted in 30, 33 and 44% attraction, respectively, in children between 3 and 5 years of age. It is difficult to predict what sort of fragrances will attract children.

The above mentioned examples of characteristics of CAP were mainly identified in studies about children’s food preferences. There are no studies, for any of the characteristics mentioned, that tested experimentally children’s ingestion likelihood with regard to different characteristics of cosmetics and liquid household products. Thus, to the best of our knowledge, there are no experimental data available that show directly that, for instance, cosmetics with a sweet smell, strong colours or cartoon characters displayed on the packaging are more likely to be ingested than others. Nevertheless, the above mentioned characteristics can serve as proxies to evaluate whether products are more or less child- appealing, until more specific data are available. More systematic research, in particular, should be done on children’s reactions to non-food products in order to better understand how children may react in front of a package and label design.

8.3. Food-resembling or child-appealing product characteristics and the probability for accidental ingestion

An overview of the characteristics that make a product relatively more food-resembling or child-appealing is given in sections 8.1 and 8.2. A ranking of the characteristics is not possible, given that there are no data available that allow for a direct comparison of the impact of the features on the risk of poisoning or ingesting the product.

However, in order to be able to better compare products and product designs with regard to their food-resembling or child-appealing properties, a simple summary score for each of the characteristics mentioned above could be obtained. A product that has a food-resembling shape, colour and smell, with a packaging that displays food-imagery, is probably more likely to be mistaken for a food than one that has only a food-resembling colour. Similarly, a product that displays cartoon characters on the package, tastes and smells sweet is probably more child-appealing than a product that just tastes sweet. However, given the limited data basis, and given that that the appeal of a product for children cannot be defined objectively, both CPRF and CAP scores would have to be interpreted cautiously and only have heuristic value until more systematic research is available.

The use of bittering agents as “aversives” has been advocated as a possible method of preventing toxic ingestions by children. The most commonly recommended agent denatonium benzoate (Bitrex) was found to have an unpleasant and bitter taste at concentrations as low as 50 ppb in liquid products (Berning et al. 1982, Hansen et al. 1993, Lawless et al. 1982, Payne 1988, Sibert and Frude 1991).

There are no published data on the effectiveness of aversing agents in limiting the ingestion of household products. Anecdotal information (Klein-Schwartz and Oderda, 1991) indicates that it may not prevent significant accidental ingestions. A single swallow of some products, such as caustics and hydrocarbons, may be toxic. Addition of aversive agents would not be effective on the outcomes of such ingestions.

Hydrocarbons are especially noteworthy because they produce toxicity by being aspired rather than by being ingested. Children may vomit after drinking denatonium benzoate spiked liquids. Thus the addition of denatonium benzoate to hydrocarbons might actually increase the potential for toxicity of these ingestions, because the act of vomiting increases the risk of aspiration.

There is no information available on effects of ingestion of products containing bittering agents by the elderly.

Source & ©: SCCS,  Opinion on the potential health risks posed by chemical consumer products resembling food
and/or having child- appealing properties, (2011), 8. Characteristics of CPRF and CAP, pp. 18-22