Food Marketing, Eating Behaviors and Gender Among Children and Adolescents: A Scoping Review

Background Pervasive marketing of unhealthy foods is one of the main drivers behind the global epidemic of childhood and adolescent overweight and obesity. Sex and gender differences come into play in the design of and responses to these marketing strategies, contributing to the perpetuation of stereotyped behavior and generating disparities in food choices and health among boys and girls. The purpose of this paper is to review the current the literature regarding gender differences in food marketing design and perception among children and adolescents to facilitate evidence-based policy dialogues to address gender-based health disparities in NCD prevention. Methods Scoping review of articles published in scientic journals in English and Spanish from 2003 on that addressed the inuence of food marketing on eating behaviors among children and adolescents including a gender perspective. The methodological quality of each article was assessed following criteria specic to each study design.


Background
There is no doubt that the current childhood overweight epidemic is one of the most serious public health issues globally. The most recent global data suggest that 5.6% of all children under the age of ve where overweight in 2019, following an upward trend over the past 20 years [1], while in 2016, 18% of the global population 5 to 18 years old suffered from the same condition [2]. Overweight causes not only many crippling physical illnesses, but also psychosocial issues, such as negative body image, reduced self-esteem, social discrimination/isolation and depression [3,4], thus hindering the comprehensive development of children and adolescents.
One of the main culprits in this trend is the in uence of food advertising on eating behaviors, particularly those that target children and promote the consumption of energy-dense, processed food products that are high in sugar, salt, and fats [5][6][7][8][9]. There is a growing body of evidence demonstrating that children's TV shows in Latin America include more food ads than programs targeting the general public; moreover, the food products marketed to children in TV have a lower nutritional quality than products advertised to other audiences [10]. This has also been documented in Argentina [11,12]. The food industry invests heavily in marketing these products, most of which do not comply with current dietary guidelines, such as sugarcoated breakfast cereals, sweets and candy, soft drinks and fast food [13]. They also employ a wide range of media and diversity of strategies to deliver its messages, including advertising in school environments, food packaging, TV, internet and social media platforms [10,14,15].
International recommendations to tackle the overweight epidemic [7,13] include the restriction of food advertising, underscoring the urgent need to implement policies that limit, or even eliminate, the exposure of children and adolescents to marketing of food products with low nutritional quality. Sex (biology) and gender (cultural) differences are relevant to overweight and obesity. Gender differences in food choices and dietary patterns are framed within social factors including societal expectations and stereotypes for males and females being transmitted via parental, peer and media in uences. A literature search conducted by Sweeting [16] showed small male and female differences in obesity rates with no main predominance of either one within particular age groups. However, the study found differences in patterning of body fat, the fat levels at which health risks become apparent, levels of resting energy expenditure and energy requirements, ability to engage in certain physical activities and the consequences of obesity for the female reproductive system. Cultural differences identi ed in the study include food choices and dietary concerns, overall physical activity levels, body satisfaction and the long-term psychosocial consequences of childhood and adolescent obesity.
In this context, it is necessary to gain better understanding of the interplay of gender and food marketing, since there is evidence that health indicators, particularly those that are intimately related to lifestyle choices -eating behaviors, physical activity, etc.-are heavily determined by gender [17]. The evidence also shows that advertising tends to offer and exacerbate traditional and stereotyped images of men and women and gender roles [18][19][20]. Advertising often contributes to consolidate gender stereotypes that perpetuate culturally rooted social and health disparities and condition differential food choices among boys and girls.
Here we present a scoping review of literature dealing with food marketing aimed at children and adolescents that also include a gender perspective. This analysis was conducted in the framework of the multi-component regional research project "Food Marketing targeted to kids: a collaborative and policyoriented study in Argentina, Bolivia, Guatemala and Peru". The overall purpose of this review was to assess the current state of the evidence regarding gender differences in the in uence, design and perception of food marketing and to identify knowledge domains that would bene t from further inquiry. The ultimate goal of this study was to facilitate evidence-based policy dialogues to reduce gender-based health disparities in NCD prevention policies.

Methods
We followed the guidelines proposed by Levac, Colquhoun and O'Malley [21] to conduct scoping reviews. We identi ed and assessed the methodological quality of published scienti c articles that addressed the in uence of food marketing on eating behaviors among children and adolescents including a gender perspective. Speci c dimensions of interest were gender differences in: a) the effect of food marketing on food intake; b) the effect of food marketing in food choice and preferences; c) responses to speci c marketing strategies and techniques; d) perceptions and attitudes towards food marketing and the need for its regulation; and e) advertising content and exposure.

Inclusion criteria
Studies were considered eligible for this scoping review if they met the following inclusion criteria: a. presented evidence pertaining to children and adolescents up to 18 years old; b. was published in Spanish or English from 2003 on, c. addressed at least one of the following themes: a theoretical discussion of the association between food marketing strategies and gender disparities; a model to estimate how food advertising incorporates gender in combination with other attributes (age, race, socioeconomic level); differential food marketing strategies to target speci c genders; an evaluation of gender-based differences in the in uence of food marketing on eating behaviors; The decision to include 18-year-olds as adolescents resulted in the incorporation of a few works that focus on young adults. Editorial and commentary pieces, "gray" literature and articles that did not report speci c data pertaining children and adolescents were excluded from this study.

Search strategy
An electronic database search was conducted using PubMed and EBSCOHost; the latter includes Business Source Premier, CAB Abstracts, EconLit, LISTA, SocINDEX, Cairn.info, Directory of Open Access Journals (DOAJ), ERIC, HeinOnline, JSTOR, OECD iLibrary, Persée, SciELO, World Bank eLibrary, MEDLINE, EMBASE and LILACS. The structure of the search was [(food marketing AND gender) OR (food marketing AND femininity) OR (food marketing AND masculinity)] AND (child OR children OR teenager OR boy OR girl).
Additional articles for potential inclusion were identi ed in a second stage by hand-searching the reference lists in relevant articles.

Data extraction
Two researchers conducted a two-stage screening process to identify articles that met all inclusion criteria.
First, titles and abstracts were analyzed to exclude clearly irrelevant articles and remove duplicates, and then the eligibility of these pre-selected articles was con rmed by evaluating the full text. Disagreements regarding inclusion/exclusion were settled by discussion between the two researchers.
The following information was collected for each included study: full reference (authors, year of publication, journal), country, study design, objectives and main results. The main themes and sub-themes explored in each paper were identi ed and a narrative synthesis was developed inductively based on these themes.

Quality assessment
The quality of each article was evaluated using different guidelines that establish the criteria to be met for each study design. Relevance, Appropriateness, Transparency, Soundness (RATS) guidelines (https://www.equator-network.org/reporting-guidelines/qualitative-research-review-guidelines-rats/) were used for qualitative studies, Effective Public Health Practice Project (EPHPP) guidelines, (https://merst.ca/wp-content/uploads/2018/02/quality-assessment-tool_2010.pdf) for experimental studies, and CEBMa for cross-sectional designs (https://www.cebma.org/wp-content/uploads/Critical-Appraisal-Questions-for-a-Cross-Sectional-Study-july-2014.pdf). Studies that analyzed advertising content were evaluated using an instrument speci cally designed for this purpose, based on the methodology proposed by Leamy and colleagues [22].

Results
Electronic searches in PubMed and EBSCOHost yielded 364 potentially relevant articles, of which 120 were duplicates and were removed from the list. Screening of title and abstracts excluded 144 articles, leaving a total of 100 potentially relevant articles of which 62 were excluded after full text assessment. The remaining 38 articles were included in this review. By analyzing the reference lists in these articles, an additional 2 were identi ed as relevant for this work, resulting in 40 articles selected for the review (Fig. 1). A total of 42 studies (Table 1) were analyzed -two articles included two separate studies each-of which 19 were experimental and 23 had descriptive, cross-sectional designs. Over half the studies (n 22) were found to have low methodological quality, while 11 and 9 were of medium and high quality, respectively ( Table 2). Boys who viewed food ads presented higher snack intake than boys exposed to neutral (non-food) ads. In turn, snack intake was slightly lower among girls when they viewed food ads than when the viewed neutral ads. Food intake decreased signi cantly with age and increased with hunger and liking of the test food products. Children who received maternal encouragement to be thin ate more when exposed to food ads (vs. non-food ads), while children that did not receive such encouragement ate more when exposed to non-food ads. No signi cant main effect was found for ad types (energy-dense food, "light" food, non-food) or gender. Exposure to food brand impressions varied among boys and girls depending on age and movie MPAA ratings. Girls were more exposed to brand impressions from G/PG-rated movies, while boys were more exposed from PG-13/Rrated movies. Girls were more likely to choose products based on color preferences and aesthetic appeal, while boys were more interested in the interactive features of the products.  Table 1 presents a summary of each study's main objective and outcomes, as well as its quality score. Table 2 illustrates the distribution of studies of low, medium and high methodological quality within each dimension of interest.
Gender differences in the effect of food marketing on food intake among children and adolescents A group of twelve studies analyzed the differential effects of food marketing on eating behaviors among children and adolescents, particularly of exposure to food advertising in different media (TV, internet, etc.), although TV was the most frequently investigated one. Most of these studies were classi ed as having low methodological quality (Table 2). Only four studies [27,29,30,35] found signi cant differences between genders, but the gender identi ed as more strongly in uenced by food marketing varied among studies.
An experimental study conducted in the Netherlands on children of 8 to 12 years of age by Anschutz et al. [30] reported that boys tended to eat a greater amount of food when viewing TV shows embedded with food ads vs. non-food ads, but this association was not found among girls. Also in the Netherlands, Anschutz et al. [35] found a higher food intake among young female university students (20 years old on average) exposed to food ads, compared to the women exposed to neutral ads. The opposite association was found in young men, who tended to consume less food when exposed to food ads vs. non-food ads.
Finally, a study conducted by Anderson et al. [27] on children and adolescents (9-14 years old) in Canada found that girls with overweight/obesity were more vulnerable to marketing techniques and increased their energy intake after exposure to food ads, while girls with normal weight and boys did not. Similar results were obtained by Keller et al. [29] in the USA to determine the impact of familiar fast-food branding on testmeal intake in children (7-9 years old). Girls were found to increase food intake in the presence of familiar fast-food brands, while boys ate similar amounts when presented with fast-food branded items vs. neutral food items (i.e. in plain white containers).
The eight remaining studies that analyzed the effect of exposure to food advertising did not observe signi cant differences between genders [23,26,29,34,39,44,46,50], although they made some noteworthy observations regarding this in uence and other intervening variables. One study found that brand recognition in TV ads of products high in sugar, fat and salt was a predictor of BMI in children as young as 3 years old [44], while other studies observed that maternal encouragement to be thin [39] and intra-family communication [50] had a moderating effect in the impact of food advertising in school-aged children. Young children with overweight were found to be more responsive to branding of food items than children with normal weight [29]. More recent studies have also explored the impact of online advertising. Exposure to online food ads was found to be associated with increased consumption of unhealthy food products [46] and the in uence of food advertising was observed to depend on the type of ad and media [26]. Finally, exposure to food ads in interaction with transportability, or the tendency to become engrossed in what one is viewing, was also found to have a signi cant effect on food intake in undergraduate university students [34].
Gender differences in the effects of advertising on preferences and choice among children and adolescents.
A group of nine studies explored these issues, of which four [17,24,55,56] generated evidence that food advertising exerts a more powerful in uence on choices and preferences among male children and adolescents than among females.
An experimental study by Chernin et al. on children aged 5 to 11 years old in the USA found that exposure to food advertising increased the children's preference for the advertised products, an effect that was more powerful among boys [24]. Also in the USA, Castonguay & Bakir [17] evaluated attitudes towards maintaining an active lifestyle among school-aged children after exposure to ads portraying physical activities vs. neutral ads. Boys manifested a greater intention to engage in physical activity when exposed to ads depicting sports, compared to girls overall and to boys who only viewed neutral ads. The authors concluded that boys are more likely to believe that the energy-dense, low quality foods depicted in these ads will help boost their physical performance than girls.
In Canada, Marquis et al. [55] evaluated how eating while watching TV affected purchase requests among school-aged children. Boys reported asking their parents to purchase the advertised products more frequently than girls, which suggests that boys were more vulnerable to food advertising than girls. Similar results were obtained by Kaur and Vohra [56] in a cross-sectional study regarding the in uence of promotional marketing techniques in retail stores among school-aged children in India. The analysis of questionnaires administered to the children's mothers showed that these promotion strategies affected boys' purchase requests more strongly than girls'.
The remaining six studies that assessed the impact of advertising on preferences and choice did not show signi cant differences between genders [25,28,37,43,52,57]. However, these studies shed light on the speci c ways in which this in uence operates. These studies did not only con rm advertising and in-store promotional strategies to be powerful in uences on food choice and preferences among children and adolescents [57], but in some cases they also identi ed other factors that moderate this effect, such as nutritional knowledge [25], the degree of attention to advertisements [28] and socioeconomic variables other than gender [52,57].
Gender differences in responses to speci c marketing techniques among children and adolescents.
Seven [31-33, 38, 54, 60, 61] of the nine studies included in this category found that male and female children and adolescents responded differently to speci c marketing techniques.
An experimental study conducted in the USA by Ogle et al. [33] focused on the in uence of depictions of licensed cartoon characters in food packaging on the attention and food preferences of school-aged children, nding that children prefer products depicting characters of their same gender. In Australia, Dixon et al. [31] also conducted an experimental study in a similar age group, where male children were observed to be more attracted to products that showed sport celebrities or that included toys, compared to girls. In an experimental study on visual memory among school children in Brazil, Gines Geraldo and Machado Pinto e Silva [38]found that girls could remember images and characters to a greater degree than boys, although no signi cant differences were found between genders in the capacity to remember colors.
Two studies focused on marketing techniques that involve active interaction with products. Using an experimental design, Hobin et al. [32] found that Canadian children were more likely to choose healthier products if toys were offered as a premium only with healthier foods, but this effect was stronger in boys than in girls. In turn, by means of focal groups, Elliot [61] explored the marketing techniques involved in "fun foods" and observed that products with enhanced interactive features were more appealing to male schoolchildren, whereas female choices were more determined by aesthetic and associative values.
As for adolescents, Vila-López and Kuster-Boluda [54] explored the importance of visual and information cues in food packages in the attitudes and preferences of adolescents in Spain. While visual cues were not found to be a signi cant in uence for either gender, information cues were observed to affect female adolescents more strongly than males, because females tended to be more worried about weight control and health overall and these factors played more heavily on their food choices.
Finally, Hattersley et al. [60] assessed gender differences in responses to the marketing of soft drinks among young Australian adults (18-30 years old). This study found that men were more in uenced by the marketing of soft drinks, while women were more strongly in uenced by the promotion of "healthier options" (e.g. fruit juice).
Of three studies that did not observe signi cant differences between genders, one found that the interaction of healthy/unhealthy slogans with healthy/unhealthy products had a signi cant in uence in purchase intent among 15-year-olds in Belgium [36]. Another study assessed the potential of using cartoon characters to promote fruit and vegetable consumption among children, with varying results by type of character and age of the audience [53]. The third paper corroborated the power of health claims and their impact on children's perceptions of relative healthfulness of food products with low nutritional quality [37].
Gender differences in perceptions and attitudes towards food marketing and marketing regulation initiatives.
This group includes ve studies [47,49,51,58,59] whose ndings suggest that children and adolescents presented some level of awareness regarding the persuasive intent behind marketing, particularly gendered advertisements, and that males and females have different perspectives regarding the need for regulation of food marketing.
Focus groups conducted by Bunting et al. [59] in New Zealand explored perceptions involving the marketing of energy drinks among adolescents and young adults (16 to 35 years old). All participants demonstrated being highly aware of the fact that energy drinks advertising targeted preferentially men or women, and that this implies a certain level of manipulation of audiences to choose products based on gendered perceptions. In contrast, a similar design implemented by Busse and Díaz [58] among school-aged Peruvian children found that all boys were aware of the persuasive intent in TV advertising, but some girls did not, suggesting a greater vulnerability to advertising persuasion in girls than in boys.
Adachi-Mejia et al. [49] assessed receptivity to TV advertising among 10 to 13 years old in the USA, in which having a favorite advertisement was used as a measure of receptivity. Although both girls and boys showed similar receptivity levels, among respondents who declared having a food ad as their favorite, boys were more likely to prefer a beer ad than girls. Conversely, girls were more likely to name a candy ad than boys.
Bhawra et al. [47] and Grunseit et al. [51] gauged the support for food marketing regulations among adolescents and young adults (16 to 30 years old) in Canada and young Australian athletes (15 to 23 years old), respectively. In both cases, women were more likely to support food marketing restrictions and related policies. Additionally, female athletes also showed a greater level of disagreement with the promotion of unhealthy foods associated to sports than men [51].
Gender differences in exposure to food advertising and gendered marketing content targeting children and adolescents.
There were six studies [17, 40-42, 45, 48] that explored either gender differences in exposure intensity or gendered marketing content. Results suggest that boys were exposed to food advertising more frequently than girls and that there is a gender bias in food marketing.
When administered questionnaires, boys reported that they viewed/listened to food ads more often and spent more time watching TV than girls, both in the USA [48] and Europe [45].
Studies focusing on advertising content found signi cant differences in both the frequency with which each gender was portrayed and the associated types of messages or foods. Castonguay and Bakir [17] observed that only 18.8% of TV ads in the USA had exclusively female characters and 6.7% showed both genders, while 74.6% showed only males. Moreover, there was an association between food healthfulness and gender, where women were more likely to be associated to healthful products and men to unhealthful ones. In the UK, Childs and Maher [40] also found that TV ads were gender bias towards boys, who were portrayed more often, either in image or as voice-over, than girls. Harrison [42] also included ethnicity in their study of gendered content in TV advertising, dividing ads into two groups: those portraying Black characters and those that did not. They observed that male characters were overrepresented in both groups (62.7% and 65.4%).
Skatrud-Mickelson et al. [41] followed a very different approach to this issue and estimated exposure by combining audience composition data, ticket sales for the top-20 box-o ce movies in the USA, and the number of times speci c food brands appeared on screen in each of these movies. Their results suggest that gender differences in exposure to advertising were determined by the interaction of age and movie age rating. Exposure among girls was greater than boys for ages 6 to 11 and 12 to 17 in G/PG rated movies, while boys of all ages had a higher exposure in PG-13 and R-rated movies.

Discussion
Children are particularly vulnerable to persuasive messages used in marketing as their cognition is relatively limited for the recognition of marketing communication [62]. Food marketing exposure in uences their preferences and, ultimately, their consumption [6,63]. There is already strong evidence that shows the linkage between food marketing and childhood obesity [64]. However, little is known about the role of gender in the promotion of products high in salt, sugar and fat to children and adolescents.
Food products do not have a gender per se, but marketing strategies are constructed in such a way that products have become gendered, resulting in foods that are more appealing to girls or to boys depending on the promotional techniques applied. Most of the studies included in this review agreed that food advertising and promotion has a similar impact on the consumption of unhealthy foods on both boys and girls. Several works found boys to be more frequently exposed to food advertising and their preferences to be more affected by this exposure, as compared to girls. This occurs in a context where advertising content presents gender bias and preferentially targets boys. The studies also bring to light the many variables that could mediate the association among advertising, gender, and eating behaviors, such as weight status, nutritional knowledge, advertising techniques, type of media, brand awareness, intra-family pressures, and others.

Strengths And Weaknesses Of The Selected Studies
The main strength of this review is that it contributes to the systematization of data regarding the role of gender in food advertising and in providing new research lines for future studies.
One limitation of this study is that the body of research reviewed is heavily biased towards TV advertising, with very few studies focusing on the impact of relatively new marketing channels, such as online games or social media platforms. Another limitation is the lack of common de nitions for key variables, such as "persuasion" or "attention". It is di cult to compare results, as there is no standardization in the measurement of outcome variables.
Methodological quality was uneven among the studies, although certain dimensions seem to have been explored using consistently more sound methodologies than others, such as gender differences in exposure to advertising and gender bias in advertising content. In contrast, most of the studies assessing the effect of marketing on food intake were of low quality. However, no associations between study quality and the identi cation of gender differences in any of the issues of interest for this review were identi ed.
Perhaps the most serious shortcoming of the work discussed in this review is the poor development of a common theoretical framework to investigate gender issues. Most of the studies used biological sex as a grouping variable, taking gender as an unproblematic construct that could be perfectly equated to sex. In order to make a meaningful contribution to gender studies, the research of the in uence of food marketing on the eating behaviors of girls and boys must account for the cultural, economic and social factors that come into play in the construction of gender identities and how these factors in uence the relationship between food advertising and people, mediated by social expectations and stereotyped "masculine" and "feminine" behaviors, attitudes, and preferences [65]. Most of the studies reviewed here have taken a simplistic approach that limits the question to biological, genetic or metabolic differences between the sexes, which also precludes any possible consideration of other, non-binary gender identities. Moreover, most of the studies were conducted in Europe, the USA or Australia, with little to no representation of the developing world, where gender disparities could be much greater than in these high-income Western countries.

Areas identi ed for future research
The body of work explored in this scoping review is a testimony to the increasing interest of the academic community in exploring gender-related determinants to eating behaviors and its implications for disparities in public health. We have identi ed issues that would particularly bene t from further inquiry, especially in the developing world, including gender stereotypes portrayed and reinforced in food advertising and how they have changed over time, how exposure to advertising varies between children and adults, the impact of new media and the role of in uencers.

Conclusions
This narrative review revealed that gender plays an important role in the development of food marketing techniques and how children respond to them. Boys and girls react in a different way to marketing, and some studies have shown that boys could be more affected, not only because they are more intensively exposed to food advertising, but also because food advertisements are usually male-oriented. Male dominance in food advertising targeting children might be a re ection of an existing cultural bias rather than an explicit decision made by food companies, but the issue merits further research, with particular attention to potential gender biases in different food categories, age sub-groups and socio-economic levels. This review also highlights the need to consider confounding variables that could potentially interact with gender and explain differential effects of advertising on food intake, including nutritional status, nutritional knowledge, type of marketing strategy and media, and previous contact or familiarity with the advertised brand, among others.
In view of the signi cance of the gender dimension in food marketing demonstrated in this review, it is imperious to include a gender perspective in policies that aim to regulate food marketing to children and adolescents. Gender is a transversal dimension that interacts and enhances all other forms of health disparities, and it must be considered in policies that address obesogenic environments.

Declarations
Ethics approval and consent to participate Not applicable.
Consent for publication Not applicable.
Availability of data and materials Not applicable.
Competing interests The authors declare that they have no competing interests.  Flow chart of the selection of studies for inclusion in the review.