Mothers' and health workers' exposure to breastmilk substitutes promotions in Abidjan, Côte d'Ivoire

Abstract Marketing of breastmilk substitutes (BMS) continues to undermine breastfeeding globally, and low income countries experiencing rapid economic growth are especially vulnerable as expanding BMS markets. The objective of the study was to understand the prevalence of exposure to BMS promotions among mothers of children 0–23 months, the frequency and type of contacts between BMS companies and health workers and the presence of educational/informational materials and branded equipment associated with such companies in health facilities in Abidjan using the World Health Organization's NetCode protocol. The methods included structured interviews with health workers and mothers and observations of equipment/materials in a sample of 42 facilities, 330 mothers and 129 health workers. Descriptive statistics were produced, and chi‐squared tests were used to assess differences by child age and facility type. Forty‐three per cent of mothers were advised to feed BMS products in the past 6 months, with a significantly higher percentage of mothers of older children (6–23 months) advised compared to infants 0–5 months. Two thirds (66%) of mothers had seen promotions outside of facilities. Among health workers, 63% were contacted by BMS companies, and only 8% were familiar with the International Code of Marketing of BMS. Differences were found between public/private facilities in the types of requests BMS companies made to health workers. Strong actions are needed in Côte d'Ivoire to prevent BMS promotion in the health system, including increasing health workers' knowledge of the International Code and national regulations, monitoring violations and reaching mothers and families to promote optimal breastfeeding practices.


| INTRODUCTION
Breastfeeding helps prevent more than 800,000 deaths of children under the age of five each year (Victora et al., 2016). In addition to preventing child morbidity and mortality and improving cognitive development, breastfeeding also reduces mothers' risk of breast and ovarian cancers and certain cardiovascular diseases (Neovita Study Group, 2016;Sankar et al., 2015;Victora et al., 2016). Exclusive breastfeeding, meaning giving infants only breastmilk and no other liquids or foods, for the first 6 months of life, is highly protective against morbidity and mortality, yet globally, only 41% of infants are exclusively breastfed from 0 to 5 months of age (Global Breastfeeding Collective, 2019;Sankar et al., 2015;World Health Organization et al., 2007). Many factors influence breastfeeding practices, but growing evidence suggests that marketing of breastmilk substitutes (BMS) has a strong influence, especially in growing low-and middleincome country (LMIC) markets (Piwoz & Huffman, 2015). BMS companies began scaling up advertising and other promotions in LMICs in the 1970s, often employing marketing tactics such as giving gifts to health workers and deploying staff dressed as nurses to provide product donations directly to mothers (Brady, 2012). As the influence of such promotions spread, health providers and researchers noted significant increases in childhood illnesses and deaths linked to formula feeding (Muller, 1974). Not only were infants missing out on the immunological benefits of breastmilk, but often, mothers mixed products improperly or with contaminated water, exposing children to diarrhoea and other infections (Jelliffe, 1972). In recent years, BMS companies' marketing tactics have expanded to reach mothers through mothers' clubs, infant feeding advice hotlines, social media, gifts and promotional materials, as well as through more traditional channels such as the media and health care providers (Hastings, Angus, Eadie, & Hunt, 2020).
In 1981, the World Health Assembly (WHA) adopted the International Code of Marketing of Breastmilk Substitutes (the Code).
Subsequent WHA resolutions provide details and guidance for its effective implementation to reduce mothers' and health workers' exposure to inappropriate BMS marketing. However, the International Code is not binding, and it is up to each member state to incorporate it into its national legislation and ensure its enforcement and monitoring. Significant gaps remain in adopting its various elements into national legislation (World Health Organization, 2020). The 2020 Status report on National Implementation of the International Code found that 136 of 194 World Health Organization (WHO) member states had enacted some Code provisions into national legislation.
However, only 79 countries prohibit all forms of BMS promotion in health facilities, where health workers play a critical role in protecting and promoting breastfeeding. Pregnant women and new mothers are particularly vulnerable to BMS marketing tactics due to their reliance on health workers as a trusted source of infant feeding advice (Haroon, Das, Salam, Imdad, & Bhutta, 2013;Sanghvi, Jimerson, Hajeebhoy, Zewale, & Nguyen, 2013). Only 30 of 194 countries prohibit BMS companies from giving any type of gifts or incentives to health workers, according to the 2020 Status Report.
In West and Central Africa, seven of 24 countries have no legal measures in place, and only three countries have legal measures that are 'substantially aligned' with the Code-the Gambia, Ghana and Nigeria (World Health Organization, 2020 year since 2012, with rising urbanization levels (The World Bank, 2019, 2020). Marketing and utilization of BMS is likely to be most common in large urban cities, where mothers often work outside the home, making exclusive breastfeeding more challenging (Barennes et al., 2012;World Health Organization & UNICEF, 2017). BMS consumption is correlated with greater household wealth in LMIC, perhaps due to its perception as modern and aspirational in some contexts (Neves et al., 2020)

Key messages
• BMS promotions at the community level and in the health system are widespread in Abidjan, Côte d'Ivoire.
• New mothers are commonly advised to feed BMS by health workers and family members and are exposed to BMS advertising through mass media.
• Strong actions are needed to prevent BMS promotions through the health system and community and increase health workers' knowledge of the Code and National BMS Decree; the implementation of a national Code monitoring and enforcement system is an important measure.  With regard to the scope of the products covered, the Code stipulates that BMS may include 'infant formula; other milk products, foods and beverages, including bottlefed complementary foods, when marketed or otherwise represented to be suitable, with or without modification, for use as a partial or total replacement of breast milk', in addition to feeding bottles and teats (World Health Organization, 1981). WHA resolution 69.9 provides guidance for all commercially produced foods that are marketed as being suitable for infants and young children 6-36 months of age and further clarifies that BMS may include any milks that are specifically marketed to feed infants and young children up to 3 years of age (World Health Organization, 2016). The scope of the National Decree is similar to that of the Code but does not take into account the recommendations of Resolution 69.9. It prohibits the marketing of BMS or other milk products, foods or drinks, including complementary foods, when they are marketed to replace breastmilk, as well as bottles, teats and similar products (Decret 2013-416 du 6 juin 2013 portant reglementation de la commercialisation des substituts du lait maternel, 2013). It is important to note the study assessed the type of feeding advice given by health workers to mothers, as well as the prevalence of interaction between BMS company representatives, which do not represent violations of the Code or the National Decree but may negatively impact breastfeeding practices. The Code and the Decree do not prohibit the dissemination of information by BMS companies to health care professionals as long as the materials are restricted to scientific and factual matters and otherwise comply with the Code. In addition, Article 6.2 of the Code states that no facility should be used to promote infant formula or other products within the scope of the Code. Neither the Code nor the Decree address health worker advice to mothers; however, Article 7.1 of the Code states that health workers should encourage and protect breastfeeding.
The study was conducted in Abidjan, the economic capital and largest city in Côte d'Ivoire. The study was a cross-sectional survey

| Participants and sampling
The sampling strategy and target sample sizes were chosen following the guidance provided in the NetCode protocol.

| Health facilities
The first stage of sampling involved drawing a sample of 33 primary health facilities and 10 secondary level maternities, as stipulated in the NetCode protocol. The Ministry of Health's Health Management Information System (DHIS2) was used to generate a list of primary and maternity facilities in Abidjan and their utilization rates for child health services (vaccination, growth monitoring and/or nutrition counselling services). Any facility offering such services was eligible for selection. Although the use of public health services has increased in recent years, it is estimated that one quarter of health services are provided by private providers (Oxford Business Group, 2020). To ensure sufficient representation of private facilities, we stratified the sampling frame by facility type. Overall, there were 110 public facilities and 35 private facilities in the sampling frame. We selected 25 public facilities and eight private facilities randomly using probability proportional to size (PPS) sampling (based on utilization rates for child health services), with one facility sampled twice resulting in a final sample of 32 primary facilities. Maternities were also stratified by the type of facility (public or private), and then, six public and four private facilities were selected, which had the highest number of deliveries of each type.

| Mothers
At each health facility, mothers of children under 6 months of age and mothers of children 6-23 months of age receiving child health services and who provided informed consent were eligible for participation. Mothers were selected using a convenience sampling strategy, wherein eligible mothers were sampled consecutively until five; mothers of children under 6 months of age and five mothers of children 6-23 months of age were interviewed in each facility. A total of 330 mothers participated. Because one facility was sampled twice, two groups of mothers from that facility participated. No mothers attending the maternity facilities participated in the study as per the NetCode protocol.

| Health workers
Three health professionals were interviewed in each facility (six from the facility that was sampled twice) and in each maternity. The director, doctor, nurse, midwife and/or health worker assigned to the paediatrics, obstetrics and gynaecology departments were eligible and were invited to participate. The first three health workers who met these criteria and were available for the interview participated, totalling 129 from the 43 primary facilities and maternities sampled.

| Data collection
Prior to data collection, study supervisors and enumerators participated in a 4-day training on research ethics and study methods and to become more familiar with the data collection tools. The questionnaire was also pilot tested in one health centre in Abidjan. Data collection took place in October 2019. The research team made appointments with the lead health professional in each of the facilities selected, conducted interviews with him or her at the designated time and then requested introductions to other health workers in the facility who met the eligibility criteria described previously. When a health worker declined to participate, the data collection team approached another health worker who met eligibility criteria. The study team then approached mothers of children 0-5 months of age and 6-23 months of age while they were waiting for well-child health services (growth monitoring and immunization) and asked them if they were willing to participate. Mothers were interviewed in a semiprivate area away from the other mothers waiting to receive services.
Data were collected electronically using the Open Data Kit (ODK) application on smartphones (Hartung et al., 2010). Three types of tools were used: a questionnaire for mothers of children 0-23 months, a questionnaire for health workers and an observation guide to assess promotional and information/education material in the health facilities. The mothers' questionnaire first asked mothers whether they had been advised to feed any milk product other than breastmilk in the past 6 months. If their response was 'yes', they were asked who made the recommendation, what type of product it was (infant formula, follow up/on formula, growing up milk, baby milk for an unspecified age range, milk not targeted for babies or a combination of milk product categories) and what brand it was. Mothers had the opportunity to list multiple products if they were advised to use them more than once in the last 6 months. Mothers were also asked if anyone advised them to feed their child any other food or drinks when the child was (or if the child is currently) under 6 months of age and the type of product recommended. Response options included complementary foods or liquid, a combination of product categories or not specified. Finally, mothers were asked if they had heard or seen any promotions at the health facility such as posters, flyers and videos or were exposed to promotional messages for any milk products, bottles or teats for children under 3 years of age through the media or in the community in the past 6 months.
The health worker questionnaire assessed whether health workers had been contacted by any personnel from companies that sell baby foods, bottles or teats in the past 6 months, and if so, how the contact was made, the reason that the company made contact and the frequency by which the company made contact. If multiple companies contacted health workers, the same information was collected for each. Health workers were also asked if they had attended any professional conferences or scientific meetings in the past 2 years sponsored by such companies, their familiarity with the International Code and national laws and regulations on BMS marketing and whether they had ever received training on infant and young child feeding (IYCF), the International Code or any national BMS marketing measures.
Any promotional and educational/information materials, as well as equipment bearing the logo of BMS companies found in the health facility, and the type of product promoted (infant formula, follow up/on formula, growing up milk, any other milk for children 0-36 months, any other food or liquid for infants, commercial complementary foods, feeding bottles or teats or not a specific product) were noted using an observation guide.

| Analysis
Data were exported from the online questionnaires and were reviewed for any outlying or missing values and checked for consistency between questionnaires. Data from the observations were entered into an Excel file. Descriptive analyses were conducted using STATA, and frequencies were evaluated for each variable.
Chi-squared tests were used to check for any statistically significant differences in BMS promotion exposure between mothers of children 0-5 and 6-23 months, as well as between mothers attending public compared to private facilities. We also conducted chi-squared tests to assess differences in health workers' exposure to BMS contacts by type of facility (public compared to private).
About half had 1 to 5 years of experience in their role.

| Mothers' exposure to BMS promotions
Data on mothers' exposure to BMS promotions both inside and outside of health facilities are displayed in Table 2. Among the mothers with infants under 6 months of age (n = 165), a time when exclusive breastfeeding is recommended by the WHO, more than half (57.6%) were advised to start feeding their infants food or drinks other than breastmilk. Almost half (43.3%) of all mothers in the sample were advised to feed BMS products in the past 6 months. There were significant differences between children's age groups, with 56.8% of mothers of children 6-23 months being advised to feed BMS, compared to 29.7% of mothers of children 0-5 months (p < 0.05). Among all the mothers, about 27% received this advice two or more times, resulting in a total of 193 instances in which mothers received advice about feeding BMS products in the past 6 months, and no significant differences in advice frequency were found between child age groups.
The advice mothers received did not differ between mothers attending public and private facilities either.
Within these 193 instances, 66.3% of the products recommended were infant formulas, and about 25% were follow on formulas. There were significant differences between the types of BMS products recommended by age group, with a much higher percentage of mothers of children 0-5 months being recommended infant formula (95.2%) compared to mothers of children 6-23 months (52.3%), and a higher percentage of mothers of children 6-23 months recommended follow-up formula (35.2%), versus 3.2% of mothers of infants 0-5 months (p < 0.05). Some of the mothers with older children were also recommended growing up milk or various combinations of milk product categories. There were no differences in types of products recommended by mothers attending public compared to private facilities. Francelait, Danone and Nestlé were commonly cited producers of the recommended products, with no variation between facility type or child age. Sources of recommendations for BMS products were health centre staff including midwives (19.0%) and paediatricians (13.7%), as well as close friends and family members (30.5%) and store/pharmacy staff (16.8%), and no differences were found between public and private facilities or child age. In the past 6 months, very few (7%) mothers had reported seeing any BMS product promotional material in the health facility. However, about two thirds (66.4%) had seen or heard BMS promotions through the media and community, including TV, radio, magazines, shop/pharmacy ads, billboards, social media, internet or community events. Among these 219 mothers reporting any exposure to promotions through the media and community, there were a total of 347 instances of exposure, because some mothers reported multiple instances and channels of exposure. The channels by which mothers were exposed included television (80.1% of instances) and stores/pharmacies (11.8%), with very few mothers reporting exposure through radio, internet and social media. About 6% of mothers had received a free sample of a BMS product, only one mother had received a coupon and no mothers had received any gifts from BMS companies. Among the mothers receiving samples, they most often reported receiving multiple samples. The sources included BMS company representatives, health professionals and retail outlets, as well as partners, parents and friends. Although the data are not displayed in Table 2, the study also assessed whether mothers had participated in any online or in-person social groups or events for mothers and, if so, if they were sponsored by BMS companies. Only 6% of mothers participated in any online or in-person social groups. Although 41% of mothers reported participating in a parenting

| Health facilities and staff
Results from interviews with health facility staff are reported in The number and type of promotional and educational material found in health facilities are shown in Table 4. We found a total of 66 BMS-branded materials in 26 health facilities. They ranged from cups (9.1%) to brochures/posters (15.2%), notebooks (7.6%), equipment such as height boards and stethoscopes (12.1%) and child clothing and feeding accessories (9.1%). Most of the materials (95.5%) were found in primary health facilities, rather than maternities. The general public was the target audience for about 64% of materials found, and health workers were the target for 27%. About one third of the materials had branding that did not focus on a particular product, but follow up/on formula (21.2%), growing up milk (15.2%) and complementary food products (15.2%) were specific types of products mentioned the most. About one quarter of products were branded by Pharmalys Laboratoire, whereas France Lait (16.7%), Nestlé (15.2%), Bledina (10.6%) and PKL (10.6%) were also common brands across these materials.

| DISCUSSION
Our study found that mothers of young children in Abidjan frequently received advice to feed BMS products to their infants and young children, including more than half of mothers of infants under 6 months of age. They often received this advice from health professionals, including midwives and paediatricians as well as members of their family and community and in pharmacies or stores. Mothers of children 6-23 months of age had a significantly higher prevalence of being advised to feed BMS than mothers with children under 6 months, but no differences were found in advice received between mothers attending public compared with private facilities. About two thirds of mothers had been exposed to BMS marketing outside of the health facility, through mass media and other advertising. More than Received a coupon for any baby milk products or feeding bottles and teats for children less than 3 years old 1 (0.6) 0 1 (0.3) Note: Results in bold indicate chi-squared test differences with p < 0.05.

EMERSON ET AL.
T A B L E 3 Contacts between companies and health facility staff (N = 129)  (Feeley et al., 2016). About 40% of mothers were exposed to any promotions outside the health facility. In Tanzania, the prevalence of exposure was quite low, with only 9% of mothers receiving recommendations from a health professional to feed BMS, and 10% were exposed to promotions inside or outside of the facility (Vitta et al., 2016). The results were comparable to several studies from Asia. In Nepal, the Assessment and Research on Child Feeding (ARCH) project found that almost half of the mothers they surveyed were recommended BMS products by health workers (Pries, Huffman, Adhikary, et al., 2016). Few mothers reported observing promotional materials in the facilities or receiving samples. In Cambodia, almost one fifth of mothers were recommended BMS products by a health professional, but 86% observed BMS promotions outside of health facilities (Pries, Huffman, Mengkheang, et al., 2016). In Latin America, one study in Mexico found that more than half of mothers surveyed had been advised to feed BMS products to their children, often by health professionals, and 80% of them had been exposed to violations  (Shrimpton et al., 2001), so mothers may be turning to BMS to assuage fears about their children's nutritional status. It is also important to note that almost one third of mothers were advised to give BMS products by friends or family members. Thus, efforts to reach these influential groups at the household and community level are needed to sensitize them to the risks of using BMS products and the benefits of practising optimal IYCF behaviours.
One of our study's limitations was that we only implemented the 'Mothers and Health Facilities' module of the NetCode protocol and did not assess violations in retail (e.g., stores and pharmacies) and the media that are also part of NetCode. We are, therefore, not able to present a complete picture of the extent of BMS promotions in  are especially vulnerable to BMS companies' marketing efforts and often lack strong systems in which product marketing can be regulated and violations monitored. As BMS companies continually adapt the channels in which they market products to have increasingly high reach, governments and health and nutrition sector partners will need to be increasingly vigilant in detecting International Code violations and protecting mothers and infants.
research and preliminary analyses. FD and RO oversaw field data collection.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.