Violations of the International Code of Marketing of Breast‐milk Substitutes in Mexico

This study estimated the prevalence of violations of the International Code of Marketing of Breast-milk Substitutes (BMS) and subsequent resolutions of the World Health Assembly (Code) at health facilities, points of sale (POS), and on BMS labelling and media in Mexico. We carried out a cross-sectional survey among 693 mothers with children aged less than 24 months and 48 health providers at public and private health facilities in two states of Mexico. Observational assessment at 20 POS and the health facilities was conducted as well as an analysis of labels on BMS products for sale. Women attending public and private health facilities reported receiving free BMS samples in the previous 6 months (11.1%), and about 80% reported seeing BMS promotion in the mass media. Health providers reported contact with BMS manufacturer representatives in the previous 6 months (15.5%), and only 41.6% of the health providers had knowledge of the Code. BMS promotions were identified at nearly all POS. Analysis of 190 BMS labels showed that 30% included pictures/text idealizing the use of BMS, and all labels incorporated health and nutrition claims. Violations of the Code are prevalent within the health services, POS, and labelling of BMS products. The high percentage of health providers with no knowledge of the Code calls for action at national level to better disseminate and comply with the Code. A transparent, free from commercial influence, and continual monitoring system for Code compliance is needed, including a follow-up component on sanctions for contraventions of the Code.


| INTRODUCTION
The contribution of breastfeeding to survival, health, and development in children as well as to the mother's health and human capital development is well known (Victora et al., 2016). Poor breastfeeding practices increase the risk of mortality and morbidity. Suboptimal breastfeeding practices have been linked to more than 800,000 deaths among children worldwide (Black et al., 2013).
Breastfeeding practices are influenced by socio-economic, cultural, and individual factors, as well as by the presence or absence of public policies that support and promote breastfeeding and regulated marketing strategies of breast milk substitutes (BMS). Among these factors, the promotion of manufactured BMS (including infant formula, follow-up formula, and growing-up or toddler milks) results in suboptimal breastfeeding practices. Promotion of BMS negatively affects the choice and ability of mothers to optimally breastfeed their children (Rollins et al., 2016). The International Code of Marketing of Breastmilk Substitutes was adopted by the World Health Assembly (WHA) in 1981 in response to a rise in child mortality owing to the promotion of BMS (World Health Organization [WHO], 1981). Since then, the WHA has adopted a number of subsequent relevant resolutions to clarify aspects of the Code. Collectively, the International Code and subsequent WHA resolutions are referred to as the Code. The Code outlines the responsibilities of governments, health care systems, and workers, as well as those of the companies that market or manufacture BMS (WHO, 1981).
The Code is a recommendation from the WHA to governments; therefore, its effectiveness in protecting mothers and families as well as health providers from inappropriate and unethical BMS marketing depends on national legislation, monitoring, and enforcement (WHO, 1981). Whereas most countries have some form of Code-related legal measures in place, in 2018, only 35 countries have legislation incorporating all or most Code provisions (WHO & UNICEF, 2018). A further challenge is the monitoring and enforcement of the Code, which are both necessary to detect and report violations and ensure mechanisms to effectively intervene, to comply with national Code-related measures. In this regard, with no change to 2018, in 2016, only 32 countries reported to WHO that they have monitoring mechanisms in place, and of those, few are fully functional in practice (WHO & UNICEF, 2016).
Even though some countries monitor compliance with the Code, the results of such monitoring are rarely published in peer-reviewed journals with broad circulation. Available studies and reports from different countries describe continued and repeated violations to the Code (Access to Nutrition Foundation, 2018; Aguayo, Ross, Kanon, & Ouedraogo, 2003;Champeny et al., 2016;Changing Markets Foundation, 2018;Ching & IBFAN-ICDC, 2017;Pries et al., 2016;Taylor, 1998).
In Mexico, breastfeeding practices are deteriorating, especially exclusive breastfeeding, particularly among indigenous women (González de Cosío, Escobar-Zaragoza, González-Castell, & Rivera-Dommarco, 2013). A recent study among Mexican women shows that social norms appeared to support breastfeeding but not as internationally recommend. Infant formula, solid food, or local beverages are commonly introduced soon after birth (Swigart et al., 2017

| Study design
The study followed the WHO/UNICEF "Protocol for the Assessment and Monitoring of 'The Code' and relevant national measurements" (NetCode protocol;WHO, UNICEF, & NetCode, 2017). This was a cross-sectional survey among mothers of children aged <24 months and health providers at public and private health facilities in the states of Chihuahua and Puebla. In addition, observational assessment was conducted at points of sale (POS) and health facilities, as well as for labels of all beverages and food products within the scope of the Code, identified in the market of the two aforementioned states.
The study was conducted during January and February 2016. The protocol was approved by the ethics committee of the National Institute of Public Health of Mexico. The Ethical Review Committee of the Pan American Health Organization, regional office of the WHO, considered the study a monitoring exercise and, therefore, not subject to human subjects review. 1 Informed consent was obtained from all study participants.

| Definitions
According to the Code, we considered the following products to be BMS: breast milk substitutes, including any kind of infant formula (for children from 0 to 6 months), follow-up formula (for children 6-12 months), and growing-up formula (for 12 months and older); other milk products, foods, and beverages, including bottle-fed complementary foods for children younger than 24 months, when marketed or otherwise represented to be suitable, with or without modification, for use as a partial or total replacement for breast milk; and feeding bottles and teats (WHO, 1981).

Key messages
• Four out of five mothers of infants under 24 months of age recalled seeing BMS publicity in the mass media.
• Eleven per cent of them reported receiving free samples of breast milk substitutes at health facilities in two cities of Mexico.
• Almost 60% of health providers lack knowledge about the Code.
• Nearly all labels of BMS products incorporated health and nutrition claims.
• It is urgent to establish actions at national level to disseminate information among health providers about the Code and government strategies currently in place to implement it. Importantly, not only monitoring of violations but also allocation of appropriate sanctions is needed.
1 "The Ethical Review Committee of the Pan American Health Organization/ WHO considered the project a monitoring exercise, and therefore, not subject to human subjects review." There are two reasons for this: first that Pan American Health Organization followed proper protocol in submitting the proposal for ethical approval and second that there are different perspectives from IRB Committees about whether Code monitoring is research in human subjects.

| States and municipalities
The sample included 10 urban municipalities, five in Chihuahua and five in Puebla, selected according to their high number of births. Puebla and Chihuahua are located in east-central and north-western Mexico, respectively. These two states were selected to be part of the study by the Mexican Secretary of Health, mainly for the following reasons: (a) These two states had different achievements in terms of actions taken to protect, promote, and support breastfeeding; (b) the health authorities of both states were willing to participate in the study; (c) the selected states were representative of two regions of the country, northern and southeast-central Mexico; and (d) there was no electoral process that might interfere with development of the study in either state.
A two-step approach was used for the selection of health facilities, mothers, and health providers, as described below.

| Health care facilities
For public health facilities, between four and five health care facilities that provided well-baby care were randomly selected in each municipality, according to the number of prenatal consults provided in the year prior to the study. Health care units were selected that attended at least 10 mothers of children aged <24 months on at least 2 days in a typical week. For private health facilities, a convenience sample was used; it would have been logistically difficult to obtain a sufficient number of women who attended them. The only two selection criteria for private health facilities were provision of well-baby care and the consent and authorization of clinic authorities for the study. The goal was to reach 75% and 25% of all public and private health facilities, respectively. Public health facilities were outpatient clinics. For private health facilities, large hospitals were also included in the sample, given the difficulty in identifying a sufficient number of women for the study from small, private, outpatient clinics. However, recruitment from private health facilities was conducted among mothers attending their child's paediatric visits.

| Mothers of infants aged <24 months and health providers
At each health care facility, 16 to 17 mothers of infants <24 months of age were identified and invited to participate in the study. Eight to nine mothers with children <6 months and the same number with children between 6 and 23 months were sampled from each selected health facility, giving a total sample size of 693 mothers in the two states. For participants with more than two children <24 months, we applied a computer-automated sampling method to randomly select only one child per mother.
During the days that we visited each health care facility, one health provider at each facility was interviewed. A total of 48 health providers from the two states were selected, based on their activities with supplying BMS in the health units as well as their availability and willingness to be interviewed.

| Points of sale
We selected supermarkets, pharmacies, convenience stores, and corner stores selling products covered under the scope of the Code. Convenience store refers to small-to medium-sized establishments that sell food products, which may be franchises or chains. Corner stores refer to locally owned establishments that are smaller than convenience stores. We visited a total of 20 POS (10 per state) close to the sampled health facilities, with three supermarkets and seven pharmacies, convenience, or corner stores visited in each state (Table 1).
Information about these establishments was obtained by asking study participants where baby milk powder or other BMS could be purchased close to the health facility.
The sample size determined was of 693 mothers, the minimum number to detect 10% of the prevalence rate of exposure to BMS promotions within the health system, with confidence intervals at 95% and a measurement error of ±5%, assuming a design effect of 2 to account for the cluster design.

| Data collection
All field workers participated in a 1-week training workshop, including a pilot study. Questionnaires were administered relative to the current Finally, advertising of BMS and other infant products was evaluated in public spaces within the health units. All pamphlets, videos, posters, growth charts, attire or materials used by health professionals, or equipment in the health unit were documented through photographs; samples of these were obtained when possible.

| Points of sale
Data collection at each POS included enumeration of products sold that were under the scope of the Code. We recorded a list of the products encountered in the sampled stores. As additional stores were visited, a master list was created by adding new products encountered that had not been found in previous stores. The brand name, subbrand name, descriptive name, and age category of each product were recorded.
For each product, the geographical location and type of business were also noted. At all selected retail outlets, any type of promotion related to BMS was recorded and photographed, and promotional materials were also collected. For analysis of BMS labels, we photographed all BMS products available at large supermarkets. We purchased products available at pharmacies, convenience stores, and corner stores, as we were not authorized to photograph products in these stores.

| Assessment of materials and product labels
A desk review was performed by one researcher (A. L. T.) to analyse the promotional and educational materials collected from health facilities and POS, as well as the labels on products sold at retail stores. All materials and labels were closely examined according to a predefined checklist of criteria that are based on the Code.

| Data management, processing, and analysis
Data were entered by field workers using Microsoft Access, 2010.
Data entry programs were designed using the questionnaires and criteria checklist to assess BMS promotions, educational materials, and labels proposed in the NetCode protocol. Data were sent to the central office of the Instituto Nacional de Salud Pública for quality control and database generation.
After a cleaning process of data out of range or data entry errors, data analysis was carried out. Means and proportions with their respective confidence intervals were calculated, taking into account the effect of the cluster sampling design (health care facilities). Statistically significant differences were tested using the chi-square (χ 2 ) test, with a two-sided P value of 0.05. All analyses were performed using Stata, version 14.1 (StataCorp, College Station, TX, USA).

| Sample characteristics
We investigated 29 public and 19 private health facilities in 10 urban municipalities of the states of Chihuahua and Puebla. No differences on maternal characteristics were found between the two states, except for education. Women from Chihuahua had higher education (Data S1). Among the private facilities, 14 were large hospitals. From the same municipalities, we visited 51 POS (35 in Chihuahua and 16 in Puebla-the field team had more opportunities to visit more POS in Chihuahua state). The most common POS visited were supermarket and pharmacies (Table 1).
A total of 693 mothers of children aged <24 months were interviewed (357 women with children <6 months); all were recruited in the waiting rooms of selected health facilities. The average age of mothers was 26 years, with more than 40% of them having completed a middle school education (Table 1). Women attending private health facilities had higher educational levels (P = 0.0001).
We interviewed 48 health professionals (one per health facility); most were physicians, and the average time working in the health facility was 8.7 ± 7.9 years (Table 1).

| Mothers of children aged <24 months
Women attending public and private health facilities reported receiving free samples of BMS in the last 6 months ( Nearly half of respondents (46.5%) reported having received a recommendation for use of BMS in the previous 6 months ( Table 2).
Nearly 30.5% of respondents received such a recommendation from physicians, followed by a relative or friend (16.5%; data not shown).
The BMS manufacturing companies that women recalled the most frequently being recommended included Nestlé (68%) and Mead Johnson (23%; data not shown).
The only difference on the Code violations between the states was those related to receiving free samples of BMS and seeing promotion/messages of BMS within health facilities and in mass media.
Mothers in Puebla reported more exposure to free samples and promotions of BMS (Data S1).

| Health providers and facilities
A total nine of 48 health providers reported having contact with BMS manufacturer representatives in the previous 6 months. Of these, only seven specified the name of BMS manufacturer (15.5%; Table 2   common in Puebla, than in Chihuahua, except for low price promotions (Data S1).

| Analysis of labelling and products
We checked the labelling and inserts of all BMS products against the  & Hernández-Ávila, 2017), we found that women reported being exposed to BMS promoted in the mass media frequently.
Producing and marketing BMS is a very profitable business.
Between 2014 and 2019, global BMS sales are projected to increase from $45 to $71 billion (Kent, 2015). The increase is mainly owing to growth in infant formula sales in low-and middle-income countries (Kent, 2015). According to EuroMonitor data provided to the WHO, in and 24 months receiving free BMS samples (Access to Nutrition Foundation, 2018; Aguayo et al., 2003;Changing Markets Foundation, 2018;Ching & IBFAN-ICDC, 2017;Taylor, 1998), in addition to health facilities receiving donations of BMS that were not used for research or professional purposes and having educational materials with BMS producer logos (Aguayo et al., 2003;Pries et al., 2016). Other studies have identified that promotion of BMS outside of health facilities is a common practice in Cambodia  and Senegal (Feeley et al., 2016), as is promotion within health facilities (Coriolis, 2014;Pries et al., 2016). BMS promotion at POS has been identified in Senegal, Cambodia, Nepal, and Tanzania . Violations of labelling standards of the Code have also been identified in other countries (Ergin et al., 2013;Parrilla-Rodriguez & Gorrín-Peralta, 2008   and Senegal (Feeley et al., 2016). As recognized in the Coriolis report Infant Formula Value Chain (Coriolis, 2014), the major global multinationals direct a large part of their sales effort towards health providers rather than retailers. The effect of a health provider's recommendation for the use of BMS on a mother's decisions about feeding practices is well known. Health providers have unparalleled credibility in terms of infant health and feeding recommendations Piwoz & Huffman, 2015). Thus, it is of paramount importance to ensure that health providers have the knowledge to adequately support breastfeeding (or know where to refer mothers who require support) and understand their ethical responsibility not to recommend the use of infant formula when there is no medical reason to do so.
Despite Mexican regulations stating that BMS manufacturers are not allowed to contact health providers, 15.5% of providers reported having contact with BMS manufacturers. Such contact with providers has also been reported in Bangladesh, Poland, South Africa, and Thailand (Coriolis, 2014), as well as in Togo and Burkina Faso . In addition, an important and disturbing finding is the lack of knowledge of the Code among health providers in Mexico, similar to health professionals in Togo and Burkina Faso . It is urgent to establish actions at a national level to disseminate the Code and develop government strategies to  There is evidence showing a potential indirect effect of the proliferation of BMS products in the market on social norms and attitudes about breastfeeding (Piwoz & Huffman, 2015). In our study, we found a great variety of products for children under 24 months old in retail, convenience, and corner stores as well as pharmacies.
Despite the Code clearly prohibiting POS promotion of BMS, this provision has not been enacted in Mexican legislation and regulations.
We identified BMS promotion in nearly all retail stores, pharmacies, Finally, an important challenge to monitoring and enforcing compliance with the Code is the increasing use of Internet platforms, cellular phones, and social media such as Facebook, Twitter, YouTube, and mobile applications to promote and sell BMS products, including online ordering and coupons for free or low-cost BMS (Abrahams, 2012). As part of this study, we explored the presence of BMS advertisements in social media, the Internet, and television programming.
The results will be presented in a separate paper, but briefly, we found a high presence of BMS promotion in all these communication media.
Thus, any strategy to establish a monitoring system in Mexico should consider the use of social media as part of the system.
Progress has been made to include some provisions of the Code in the legislation and regulations in Mexico; however, gaps remain that must be addressed. It is important to update Mexican regulations regarding follow-up and growing-up formulas; in addition, as stated in the 2016 WHA 69.9 resolution, it is also necessary to include the definition of foods for infants and young children, namely, "all commercially produced food or beverage products that are specifically marketed as suitable for feeding children up to 36 months of age" (WHO, 2016). WHA 69.9 resolution states that products that function as BMS should not be promoted.
In addition, specific implementation actions are needed for effective education, training, and monitoring systems to ensure that health In addition, an effective, transparent, and free from commercial influence national monitoring system of compliance implemented by the Government of Mexico with Code is also needed, which includes a component to follow-up on sanctions for contraventions of the Code.
integrity of the data and accuracy of the data analysis. SHC, ALLT, TSL, CKL, TGC, PS, and LGS designed the research study. SHC, ALLT, and TSL conducted the research. SHC, ALLT, TSL, CKL, TGC, PS, and LGS proposed and approved modifications to the protocol during the development of the study. SHC and ALLT analysed the data. SHC and ALLT wrote the manuscript and had primary responsibility for the final content. TSL, CKL, PS, TGC, JRD, and LGS gave key insights for data interpretation and for the final manuscript. All authors read and approved the final version of the manuscript.