Prevalence, duration, and content of television advertisements for breast milk substitutes and commercially produced complementary foods in Phnom Penh, Cambodia and Dakar, Senegal

Abstract Promotion of breast milk substitutes (BMS) and inappropriate marketing of commercially produced complementary foods (CPCF), including through television, can negatively influence infant and young child feeding. The World Health Organization International Code of Marketing of Breast‐milk Substitutes and subsequent relevant World Health Assembly (WHA) resolutions prohibit such advertising and require manufacturers and distributors to comply with its provisions; however, such regulations at national level may vary. Advertisements require Ministry of Health approval in Cambodia but are not regulated in Senegal. Television stations were monitored for 13 months in Phnom Penh and for 3 months in Dakar to assess advertisements for BMS and CPCF. Ten television channels (out of 16) in Phnom Penh and four (out of 20) in Dakar aired advertisements for BMS. Three and five channels, respectively, aired advertisements for CPCF. All BMS advertised in Phnom Penh were for children over 1 year of age. BMS products for children 6+ months of age and 1+ years of age were advertised in Dakar. Average air time for BMS advertisements was 189.5 min per month in Phnom Penh and 29.7 min in Dakar. Air time for CPCF advertisements averaged 3.2 min per month and 13.6 min, respectively. Fewer than half of BMS advertisements and three quarters of CPCF advertisements explicitly stated an age of use for products. Nutrition and health claims were common across BMS advertisements. This study illustrates the need to adopt, regulate, monitor, and enforce legislation prohibiting BMS promotion, as well as to implement regulations to prevent inappropriate promotion of CPCF.


| INTRODUCTION
Promotion of breast milk substitutes (BMS) has been shown to negatively affect breastfeeding practices throughout the world (Piwoz & Huffman, 2015;Rollins et al., 2016;Smith & Blake, 2013). This is especially a concern in low-and middle-income countries due to the increased risk of infant deaths caused by suboptimal breastfeeding practices . It has been estimated that not breastfeeding is associated with lower intelligence and economic losses of about US $302 billion annually; yet sales of BMS in 2014 were estimated at US$44·8 billion, and the market value is projected to reach US $70.6 billion by 2019 . Television advertising is likely an influential marketing component driving such sales (Kaplan & Graff, 2008). However, there is limited information on the frequency and duration of such advertisements nor is information readily available on which manufactures are advertising, which specific products are advertised, or what messages are included in the advertisements. Information on the frequency, duration, and content of television advertisements for commercially produced complementary foods (CPCF) is also not readily available.
The International Code of Marketing of Breast-milk Substitutes and subsequent relevant World Health Assembly (WHA) resolutions prohibits the promotion of BMS (WHO, 1981), including through television advertising. In 2016, the WHA resolution 69.9 urged countries "to take all necessary measures in the interest of public health to end the inappropriate promotion of foods for infants and young children, including, in particular, implementation of the guidance recommendations" . This resolution refers to the World Health Organization Guidance on Ending the Inappropriate Promotion of Foods for Infants and Young Children (World Health Organization, 2016), which was developed in response to increasing evidence that the promotion of BMS and some commercial foods for infants and young children undermines optimal infant and young child feeding (WHO, 2015). This guidance clarifies that milks for children up to age 3 years are BMS and should not be advertised.
The inappropriate promotion of foods for infants and young children has been an ongoing concern among member states at the WHA (World Health Assembly, 2012, 2016. Appropriately formulated CPCF may have the potential to improve nutrient adequacy in diets of children after 6 months of age, contributing to fill the gap when breast milk alone is no longer sufficient to cover all nutrition needs due to rapid growth (Michaelsen, Grummer-Strawn, & Bégin, 2017). However, introduction of complementary feeding before 6 months of age can displace breastfeeding (Pan American Health Organization & World Health Organization, 2003) and may contribute to an increased risk of childhood overweight (Pearce, Taylor, & Langley-evans, 2013;Pluymen et al., 2018). International regulations and recommendations have consistently emphasized the use of a variety of locally available foods for complementary feeding and have warned against marketing of commercial foods which may discourage this (Clark, Shrimpton, & Feeding, 2000). According to the World Health Organization's Guidance on Ending the Inappropriate Promotion of Foods for Infants and Young Children (World Health Organization, 2016), messages used to market foods for infants and young children should support optimal feeding. These messages should include a statement on the importance of continued breastfeeding for up to 2 years or beyond and should specify the appropriate age of introduction of the food. Furthermore, messages should not suggest the use of CPCF for infants under the age of 6 months, make a comparison or claim equivalence to breast milk, recommend or promote bottle feeding, or convey an endorsement.
In many low-and middle-income countries, it is common for most mothers breastfeed their children at some point. This is the case in record and save television programming, so monitoring took place prospectively for the three-month period.
All 13 local channels in Phnom Penh and three out of 13 international cable channels aired in Cambodia were included in this study (n = 16). The three cable channels were chosen because their content was aimed at the entire family, and they were considered attractive to young mothers. All channels were broadcast in Khmer. Six out of 12 local channels in Dakar were identified as having viewership by at least 80% of Senegalese households and so were included in the monitoring.
Out of the 30 most-watched international cable channels in Dakar, an additional 14 channels were selected for monitoring as the media firm experts suggested they were likely to be watched by mothers of young children (n = 20). Of these 14 international channels, 10 were broadcast from France, two from Mali, one from Ivory Coast, and one from Cameroon. In Dakar, international channels were broadcast in French.

| Review of advertisements
Researchers met with both firms to explain the categories of BMS and CPCF. Trained monitors in both countries reviewed the daily television programming and identified all BMS and CPCF advertisements.
BMS were defined as any formula, milk, or milk-like product labelled or marketed as suitable for feeding children younger than 36 months of age (World Health Organization, 2016). CPCF were classified as any foods or beverages (excluding BMS) recommended for introduction at an age less than 24 months of age . All video recordings of advertisements identified by the firms during monitoring were also reviewed by the researchers to ensure that products were appropriately classified.
An advertisement was included in the study if it advertised either BMS or CPCF and aired during the data collection period. Individual advertisements could be shown one or more times each day.
Advertisements were reviewed and information logged on the date, time of day, duration (in seconds), product category, manufacturer and brand. A transcript of the recording was linked to each advertisement. Each different advertisement was transcribed in Khmer or French and then translated to English for assessment of its content by the researchers. All advertisements were observed and transcripts reviewed to assess which products were included in the advertisements and to determine if suggested ages of use for the products were provided either visually or verbally. Research has shown that different manufacturers use numbered "stages" to refer to different recommended ages of use for BMS products, but that these are not consistent across manufacturers ). An advertisement (or product image shown in the advertisement) which visually or verbally stated a stage of use (a number such as 1, 2, or 3), was not considered to be a suggested age of use unless an age (i.e., "6-12 months" or "from 6 months of age") was also visually or verbally stated.
The content of advertisements was assessed by coding whether they verbally or in text made nutrition and health claims as defined by the Codex Alimentarius Guidelines for Use of Nutrition and Health Claims CAC/GL 23-1997(Codex Alimentarius Comission, 2004. Nutrition claims are broadly described as any representation which states, suggest or implies that a food has particular nutritional properties. They include nutrient content claims (defined as a nutrition claim that describes the level of a nutrient contained in a food), nutrient comparative claims (a claim that compares the nutrient levels and/or energy value of two or more foods), and nonaddition claims (any claim that an ingredient has not been added to a food, either directly or indirectly). The Codex Alimentarius guidelines define health claims as any representation that states, suggests, or implies that a relationship exists between a food or a constituent of that food and health. Health claims include nutrient function claims (a nutrition claim which describes the physiological role of the nutrient in growth, development and normal functions of the body), other function claims (which concern specific beneficial effects of the consumption of foods or their constituents, in the context of the total diet on normal functions or biological activities of the body) and reduction of disease risk claims (relating the consumption of a food or food constituent, in the context of the total diet, to the reduced risk of developing a disease or healthrelated condition). All spoken and written text in each advertisement was reviewed against these criteria; all nutrition and health claims were recorded and the total number of advertisements with each claim type was determined. Because only international cable channels aired advertisements in Dakar, no information is available on costs specific to Senegal because the fee schedule for these channels included the total price for an advertiser to air in all countries reached by the channel. Table 1 shows that all the advertisements for BMS products in Phnom

| Advertising of BMS
Penh were for formula/milk for young children aged 1+ years except for the five advertisements aired by Dumex/Danone to support their brand after a product recall which took place in 2013 (Narim & Lewis, 2013). In Dakar, products for older infants 6+ months of age were also advertised. There were no BMS advertisements in either setting for products for infants younger than 6 months.
There were 11 BMS advertisements for specific brands/products in Phnom Penh and five in Dakar (Table 2).
Two BMS advertisements out of the five observed in Dakar verbally or visually gave the age of introduction for the product. Three of the 11 advertisements for specific BMS products did so in Phnom Penh; although numbered stages used to differentiate products were shown on the labels of all but one product (Abbott Pediasure), the associated ages were not visible. One label (Friesland Campina Dutch Lady) showed 1, 2, 3 on the label with no indication in the advertisement what these numbers indicated.
Most advertisements for specific BMS products (11 out of 16 advertisements in both sites) made nutrition claims and nearly all (13 out of 16 advertisements) made health claims (Table 2). Table 3  "… it contains essential nutrients including a combination of iron and essential fatty acids to support their brain development." (BMS advertisement from Dakar).

| Advertising of CPCF
There were four manufacturers whose CPCF were advertised. Two companies advertised products in Phnom Penh and three in Dakar, with one (Nestlé) advertising in both (Table 2).
Of seven advertisements, five (two in Phnom Penh and three in Dakar) indicated the recommended age of use of 6+ months verbally or visually in the advertisement. The number 6 was shown on one product label but no other information (such as the term "months") was legible. One product label (Babybio) did not show any numbers or ages. Three advertisements had nutrition claims and four had health claims ( Table 3). Content of these ads sometimes combined nutrition and health claims: See Data S1 for English language transcriptions of all advertisements observed in this study.

| Costs
During the 13-month period monitored, the average monthly cost of Out of seven advertisements for CPCF, five indicated suggested ages for use of the product. In order to ensure optimal infant and young child feeding and protect, promote and support exclusive breastfeeding to 6 months, the World Health Organization states that messages used to market complementary foods "should specify the appropriate age of introduction of the food (not less than 6 months)" (World Health Organization, 2016). The findings of this study show BMS products for the 6+ age group, often branded as "follow-up formula" or "growing-up milk" , commonly have similar packaging and design to infant formula for infants less than 6 months of age. Advertising of these products may implicitly promote a manufacturer's BMS for the less than 6 months age group (World Health Organization, 2016). Evidence shows that caregivers in many settings struggle to correctly distinguish between similarly branded BMS products (Berry, Jones, & Iverson, 2010;Cattaneo et al., 2015), which can lead to inappropriate feeding. The use of similar branding, pictures, and logos on BMS across a wide age range has been described as detrimental to the effectiveness of national restrictions on BMS advertising (Vinje et al., 2017).   Cambodia, and Vietnam in 2015-2016(Vinje et al., 2017, with numerous products advertised and a high frequency of these advertisements. In Indonesia and Thailand, products for young children 12+ months were observed in advertisements, whereas in Myanmar, Cambodia, and Vietnam, they were for young children 24+ months. Two systematic assessments of advertising in Java, Indonesia, found television advertisements for BMS and found that television was the most commonly reported source of promotion for infant and young child foods by mothers (Durako, Thompson, Diallo, 2016a;Hidayana, Februhartanty, & Parady, 2016). A survey in Vietnam showed similar findings, with all television advertisements featuring products for young children over 24 months of age (Durako, Thompson, Diallo, 2016b). The authors raised the concern that women may be familiarized with the names of the BMS manufacturers and their brands through the advertisements for milk products for children aged 2+ years that use the same brand name and colours and designs on labels of products for those less than 2 years of age. Little infor-  vignettes, or other forms of communication anywhere on the package that characterized the product, suggested how to prepare or use it, or provided expert, health, nutrition, or ingredient information" (Pomeranz, Romo Palafox, & Harris, 2018). Packages of products for children 9-24 months and 12-36 months were found to have an average of 3.8 and 4.2 claims each, respectively. A 2017 survey of 25 infant formula websites in Australia reviewed content for the presence of health claims, nutrition content claims, and references to breast milk. These researchers found such claims to be ubiquitous, despite being prohibited by national regulations (Berry & Gribble, 2017). These expanded criteria for identifying claims on infant and young child feeding products may allow for more nuanced descriptions of the messaging used by companies to induce purchase of their products.

| CONCLUSIONS
In order to address the major public health problem of infant and young child illness and malnutrition, it is important that countries adopt, monitor, regulate, and enforce restrictions on marketing of all forms of BMS including through television. The recent World Health Organization Guidance makes it clear that a BMS "includes any milks in either liquid or powdered form, that are specifically marketed for feeding infants and young children up to the age of 36 months (including follow-up formula and growing-up milks)." It is essential that products clearly state the suggested earliest age of use and that this should not be less than 6 months for some BMS and all CPCF. In