How the marketing practices of commercial milk formula companies impact infant breastfeeding practices in China

Introduction

As recommended by the WHO and UNICEF, optimal breast feeding is to initiate breast feeding within the first hour after birth, ensure exclusive breast feeding for 6 months, and continue breast feeding up to 2 years of age or beyond with diverse complementary foods introduced after 6 months.1 Breast milk has wide-ranging benefits, including decreased morbidity and mortality from infections, lowered risk of overweight and obesity, improved cognitive development and decreased maternal deaths due to breast and ovarian cancers and type 2 diabetes.2 3 In addition, these benefits may even include alleviating the suboptimal development for prenatal deprivations among a large subpopulation, namely small vulnerable newborn.4 Globally, the small vulnerable newborns such as preterm non-small for gestational age (SGA), term SGA and preterm SGA had an extremely high prevalence of 26.2% (beyond 35 million births) according to the recent estimate in 2020, majority of which were located in low- and middle-income countries (LMICs).5 Besides, these small vulnerable newborns expect to be less likely to be breastfed. Although China has a relatively low prevalence of adverse birth outcomes as compared with other LMICs, for example, preterm rate 6.9% in China versus 10.4% in Asia, the absolute number is large and ranks second globally,6 resulting in a serious public health issue of suboptimal development. Therefore, improving breast feeding would have profound impacts on maternal health and children long-term development among generally healthy population in direct and indirect influences.

However, suboptimal breastfeeding practices are widely prevalent across the world. Globally, 60% of babies are not breastfed in the first hour, and almost 60% of infants are not exclusively breastfed, contributing to more than 800 000 child deaths annually.7 8 Moreover, many infants aged 0–23 months are fed, at least in part, using commercial milk formula (CMF).9 A recent analysis using national data from 126 countries showed that each additional kilogram of CMF sold per child annually was associated with 1.9 (95% CI 1.5 to 2.2) percentage points lower in the breast feeding rate at age 1.10 Globally, the sale of CMF was estimated to have a market value of US$ 70.6 billion in 2019, with a stable compound annual growth rate of 10.6% until 2026.8 Of the CMF sold, almost 40% were consumed by infants aged 0–6 months.11 As the population size and consumer purchasing power increase with the increasing household income in underdeveloped settings, the growth rate is expected to be even higher.12 13

The most recent data from China indicated that the exclusive breastfeeding rate for infants under the age of 6 months was 34.1%, with notable differences among provinces,14 15 and far below the national target of achieving 50% of exclusive breastfeeding rate by 2025. The CMF market in China totaled 683 000 tonnes in sales and was valued at US$ 25.97 million in 2019. In 2020, the CMF market in China represented almost 50% of the global CMF market,16 and it is projected to grow considerably over the foreseeable future.17

To limit the impact of CMF marketing on breast feeding, WHO with the support of Member States developed The International Code of Marketing of Breast Milk Substitutes and its subsequent resolutions (hereinafter the Code) for regulating inappropriate marketing and promotion of CMF. However, the weak adoption of the Code in countries, particularly in countries like China that have no administration rules of the Code, resulted in significant violations by CMF manufacturers.18 19 For example, a 2012 survey covering six cities in China reported that 40.2% of mothers interviewed received free formula samples, with a majority of them receiving the free samples in or near hospitals.20 Furthermore, CMF companies are implementing aggressive marketing strategy of digital techniques to continuously and directly reach pregnant women and mothers usually with personalised CMF promotions and advertisements at vulnerable moments in their lives, which, however, is often not recognisable as advertising and, thus, can evade scrutiny from health authorities.21 A cross-sectional survey conducted in Mexico in 2021 quantitatively showed that more than 93% of parents reported exposure to digital CMF marketing and that parents seeing higher number of digital adds had lower likelihood of exclusively breast feeding their children and higher risk of feeding CMF.22 Taken together, as indicated in the 2023, breastfeeding series in the Lancet, the marketing and distribution practices of CMF manufacturers have been a major deterrent for optimal breastfeeding practices.10 However, the pathways remain unclear as to how the specific types of CMF marketing practices, including online message dissemination and active outreach to mothers, result in suboptimal breastfeeding.

This paper seeks to elucidate the pathways that connect marketing messages, advertisements and promotions from CMF companies to maternal decisions related to infant and young child breastfeeding practices. With support from UNICEF China, a cross-sectional survey was conducted that mapped the marketing of CMF across Beijing and Jinan in 2020. Data from quantitative implementation in this survey were analysed where marketing practices of CMF companies were assessed through mothers and pregnant women reporting their exposures. We aimed to examine the associations between the marketing practices of CMF companies and infant and young child breastfeeding practices determine the influencing pathways for breast feeding, how it changes the perceptions and attitudes of mothers towards formula and breast feeding, and how it impacts breastfeeding outcomes.

Methods

A cross-sectional survey was conducted in Beijing and Jinan in 2020, two relatively developed cities in China, through face-to-face interviews using a tablet assisted software (CAPI) delivered by a trained team. As part of a multicountry study on CMF marketing practices and infant breast feeding, which aimed to create an ecology of formula feeding and mapped the marketing of CMF, the procedure details and study protocol were described elsewhere.23 Briefly, within each country, a primary city that typically functions as the centre for government, commerce and culture and is broadly representative of the country as a whole and a secondary city that differs from the primary city in aspects of size, location and sociodemography were selected. To map the CMF marketing of participants at different socioeconomic environment exposed to, the final samples were ensured to be equally distributed by socioeconomic status (eg, low, medium and high) based on their household annual income (see online supplemental table 1) in each city. Then, convenience/non-probability sampling was employed to recruit mothers with infants and young children under the age of 18 months as well as pregnant women (see online supplemental figure 1). Mothers or pregnant women under 18 years old, or having major health issues that may affect infant feeding, were excluded. In addition, for mothers who had multiple children, information on breastfeeding practices was collected only for the youngest child. After locally adopting the main protocol and standard training in China, a pilot was conducted within 100 respondents to test the full range of survey activities and help to reduce the risk of non-response. Finally, participants were recruited in local clinics and healthcare facilities and/or street-based such as in shops, supermarkets, department stores and health stores. The snowball sampling technique was also used.

Written informed consent was obtained from all participants after explaining the purpose of the survey. In addition, all participants were informed that full confidentiality would be maintained and that they could stop participating at any time without explanation.

Mothers were interviewed about their current breastfeeding practices for their youngest child. Specifically, mothers were asked by ‘how are you currently feeding your youngest baby?’, with possible answers of (1) breast milk only since birth, (2) formula only since birth, (3) both breast feeding and formula feeding from birth, (4) breastfed first and now I am formula feeding, (5) breastfed first and now I am breast feeding and giving formula, (6) breast feeding and formula feeding first and now I am giving formula and (7) breast feeding and formula feeding first and now I am breast feeding. Two categorical outcomes were derived using the information above. Predominant breast feeding was defined as infants born to mothers who answered breast feeding only from birth, and any breast feeding was accordingly defined as those who received some breast milk.

For women who were pregnant at the time of the interview, information on their intention for breast feeding during the first 2 weeks after birth was similarly collected using the similar set of questions and answers above. Pregnant women who answered breastfeeding only were defined as those who intended to exclusively breastfeed.

In addition, the mothers’ and pregnant women’s attitude towards the benefits and use of breast milk and CMF was also captured using a set of five statements: (1) formula-fed babies have improved growth compared with breastfed babies, (2) breast feeding and formula feeding provide a baby with the same health benefits, (3) formula helps babies sleep better, (4) formula is very similar to breast milk and (5) formula keeps babies fuller for longer. A principal component analysis was then applied to these questions and the factor score was used to assess maternal attitude towards the use of formula milk, with higher scores indicating a positive attitude or perception towards CMF feeding. The scores obtained were further categorised into mild, medium and strong attitudes by tertiles.

Furthermore, the survey interviewed pregnant women and mothers about possible marketing practices of CMF companies and existing formula types and brands in markets that they might expose to, without exact time limit to recall the exposure history. In the present study, we reviewed the relevant marketing practices of CMF companies and combined similar practices into several categorises, which were summarised below and in online supplemental table 2.

Statistical analysis

Numbers and percentages were used to describe the background characteristics of participants. A logistic regression was performed to examine the associations between the marketing practices of CMF companies and breastfeeding outcomes, with an adjusted OR and its 95% CI. For mothers, predominant breast feeding and any breast feeding were taken as the primary outcomes; for pregnant women, plan to exclusively breastfeed during the first 2 weeks of life was taken as the only breastfeeding outcome. The associations between scores of maternal attitude towards formula and breastfeeding outcomes were examined using the same statistical methods above. Furthermore, generalised linear models were conducted to examine the associations between marketing practices of CMF companies and continuous scores of maternal attitude towards formula with an identity link among mothers and pregnant women. The adjusted mean differences and their 95% CIs were estimated. The CMF marketing practices were mutually adjusted in the models with other covariables above.

In addition, a path analysis was conducted to explore the influencing pathways between the marketing practices of CMF companies, maternal attitude towards formula and breastfeeding outcomes. As an extension of multiple regression and a precursor to structural equation modelling without latent variables, path analysis helps to discern and assess the relative strength of direct and indirect associations of a set of variables acting on a specified outcome via multiple pathways within a hypothesised causal system, which are presented by straight arrows between variables only pointing in one direction namely no feedback loops.24 The conceptual framework of the path analysis in the present study is presented in a directed acyclic graph in figure 1. The relative contribution of each factor to breastfeeding outcomes was calculated as the percentage of its absolute total effect divided by the sum of all factors’ absolute total effect. The total effect of each factor was calculated by the sum of direct and indirect effects, which were estimated by the standardised beta of each arrow link in figure 1. All these analyses were conducted in STATA V.15.0, with a two-sided p value less than 0.05 considered statistically significant.

Request permission

Figure 1

The conceptual framework of the path analysis. CMF, commercial milk formula.

Results

Discussion

It was found that five out of the six listed CMF marketing practices had direct effects on reducing the likelihood of infants being predominantly breastfed, including (1) social media messaging on CMF, (2) online engagement with CMF companies, (3) CMF promotions and discounts, (5) feeding suggestions from CMF salespersons and (6) free formula samples distributed in hospitals, although only (2) and (3) reached statistical significance probably due to a small sample size. Besides, higher number of CMF marketing practices mothers exposed to were linearly associated with lower likelihood of predominant breast feeding, suggesting a dose–response relationship. Moreover, two out of the six listed CMF marketing practices, including (2) online engagement with CMF companies and (6) free formula samples distributed in hospitals, had indirect effects on suboptimal breastfeeding outcomes, which was partly mediated by positive maternal attitude towards formula. In addition, the impact of these influencing factors differed between mothers and pregnant women.

The results suggested that healthcare providers and the health system remained major channels used by CMF companies to influence breastfeeding decisions in China. Liu and colleagues conducted a cross-sectional survey in six cities of China in 2012 and reported that 40.2% of mothers received free formula samples in hospitals.20 The present study showed a substantially decreased rate (18%), which could be partly explained by actions that have been taken by the health system in China. The National Health Commission (NHC) has been enforcing the implementation of the Baby Friendly Hospital Initiative in Chinese hospitals and health facilities in recent years. Moreover, the NHC just published the Breastfeeding Promotion Action Plan (2021–2025), which prohibits the provision and promotion of formula in healthcare facilities and disallows CMF companies to engage with health workers. These policies are expected to reduce and limit the impact of CMF marketing on undermining optimal breast feeding. However, more actions can be taken to decrease the prevalence of mothers receiving free formula samples in hospitals. Therefore, further efforts from health-related government sectors are necessary to address the content and marketing channels used by CMF companies to actively reach and engage mothers.25

In addition, actions from other government sectors are also needed to regulate the marketing practices of CMF companies (eg, particularly 2 and 3) that directly influence infant breastfeeding outcomes. In agreement with the findings of this study, prior qualitative interviews concluded that CMF companies often host a wide range of offline events to actively reach mothers such as baby clubs, roadshows and seminars,26 effectively building trusting relationships with mothers. Besides, these activities’ or events’ information is nowadays personalised and directly sent to mothers and pregnant women by CMF companies employing marketing strategy of advanced digit technology.27 Furthermore, to garner more trust from mothers, CMF salespersons, who are not qualified health professionals, are actively providing infant and young child feeding knowledge and recommendations that are medically unsubstantiated. According to the theories and models of behaviour change drawn from psychology and informed by economics and sociology,28 29 these tactics of personally interacting with mothers to establish trust are more likely to change the health-related behaviours compared with disseminating online messages on CMF. Notably, all these marketing practices are prohibited by key provisions of the Code.30 The Code is a landmark policy framework designed to stop commercial interests from damaging breastfeeding rates and endangering the health and nutrition of the world’s youngest inhabitants, which is adopted by the World Health Assembly in 1981 as a set of recommendations to regulate the CMF marketing, feeding bottles and teats. The Code aims to ensure that CMFs are available when needed but not promoted. It points out that, given the special vulnerability of infants and the risks involved in inappropriate breastfeeding practices, usual marketing practices are unsuitable for CMF. The Code includes a number of provisions about the role of health workers and health systems and points out that health workers should make themselves familiar with their responsibilities under the Code. However, in China, some provisions in the Code, such as regulation of marketing and sale of formula in shopping malls or supermarkets, are not strictly enforced, and there are no legal sanctions for violating these provisions, especially after the abolishment of the Measures for the Administration of the Sale of Breast-milk Substitutes in 2017. To date, robust monitoring and enforcement of legal measures mainly focus on the safety and labelling of formula milk in foods and advertisements in China.31 Therefore, corresponding market regulations and measures from the government should be developed and strictly enforced in China to prohibit CMF companies from actively reaching women through promotions and discounts, such as sales promotions, discount coupons, redeemable points in stores, and free gifts.

The results of the path analysis supported the hypothesis that CMF marketing practices suggested to mothers that formula is a suitable and even better option for feeding their infants than their own breast milk, thus influencing their breastfeeding practices. Specifically, the indirect effects of two marketing practices (eg, (2) online engagement with CMF companies and (6) free formula samples distributed in hospitals) on suboptimal breast feeding were statistically significance. Moreover, among the marketing practices examined in this study, those involving interactive and active communication with mothers and pregnant women (eg, (2) online engagement with CMF companies, (3) CMF promotions and discounts and (6) free formula samples distributed in hospitals) ranked at the top for impacting optimal breast feeding among mothers. In a cohort of 695 Chinese mothers, Tang and colleagues reported that 38% thought that infants fed by formula slept for a longer duration at night than those who were breastfed.32 Similarly, a qualitative analysis also reported that CMF companies tried to use multifaceted marketing strategies and practices to demonstrate that formula was an appropriate substitute of breast milk, and this in turn reduced mothers’ confidence in their ability to breastfeed.33 34 These unethical and multifaceted marketing practices largely undermine mothers’ confidence in breast feeding, even though many pregnant women have the desire or intention to breastfeed.26 35 In order to protect and mitigate the indirect consequence of CMF marketing practices on infant and young child breastfeeding practices, interventions like breastfeeding education and counselling by nurses and health workers could be implemented and scaled up to improve pregnant women’s and mothers’ feeding knowledge and skills, thus improving optimal breast feeding.36 Of note, the Code clearly prohibits any promotions to induce sales directly to the consumer. In addition, although the legal measures in foods and advertisements prohibit the promotion of formula for infants younger than 12 months, violations are widely prevalent, especially in shops in rural areas of China. Furthermore, CMF companies use manipulative marketing tactics to promote formula for children older than 12 months, including the use of cross-promotion by having similar packaging design to CMF for younger infants,10 indirectly influencing optimal exclusive breast feeding. Therefore, the government and relevant entities should develop stricter legislation, regulatory and enforcement frameworks in line with the Code, and strengthen the implementation of existing legal measures to remove structural barriers and comprehensively regulate the marketing practices of CMF companies, in order to support optimal breast feeding.

Among pregnant women, maternal education was identified as another strong influencing factor, with higher maternal education level associated with higher likelihood of planning to exclusively breastfeed in the first 2 weeks after birth, which is in line with the findings from another study.37 This may be due to the fact that pregnant women with higher education level generally have better feeding knowledge. Interestingly, the positive influence of maternal education on optimal breast feeding substantially decreased after giving birth. From the aspect of total effects, the socioeconomic factors such as maternal age, education and occupation had relatively strong influence on infant breastfeeding outcomes, of which, of note, were partly functioned by influencing their exposures to CMF marketing practices and positive attitude towards formula. Taken together, these findings suggest that it is necessary to establish a sustainable and favourable environment for breast feeding at the individual, family and social environment levels during different life stages,38 as depicted in the 2023 Lancet breastfeeding series.39 For example, more investment should be put towards support mothers and families, such as developing and implementing family-friendly policies in the workplace (eg, paid maternal leave, breastfeeding breaks, flexible working hours, and establishment of breastfeeding rooms).

This study had limitations. First, it surveyed two cities that were relatively developed, which may limit the generalisability of the results. However, it should be noted that the trends and values on infant and young child feeding practices are generally established in urban populations and then spread to rural communities.26 Besides, the sampling method of non-probability may create potential reporting and selection bias. Second, although the marketing practices of CMF companies were collected from multiple domains, the study did not consider all the relevant marketing practices, such as the labelling of formula.40 In particular, WHO and UNICEF recommend standardised and plain packaging for formula products as part of the regulatory measures.26 Furthermore, these marketing practices were collected by interviewing what mothers and pregnant women had been exposed to, which may arise the recall bias. However, this risk should be minimal as the study subjects were mothers with young children and pregnant women, all of whom were personally experiencing the CMF marketing practices. In addition, the multiplicative interactions among these CMF marketing practices are not considered in the present study, which needs a larger sample size. Nevertheless, the additive interactions among them were presented by our results (table 3) to some extent. Third, as with other observational cross-sectional studies, unmeasured covariables/confounders like family support and temporal relationships are always concerns for confirming causal inference and thus some terms we used, for example, impact, indicate associations but not causal link. In a larger sample, some sensitivity analyses like E-value approach could be pursued to assess the impact of unmeasured confounders on result robustness.41 Future studies with strict cohort or quasi-experiment design should be conducted for confirming the causality.

In summary, the study found that the marketing practices of CMF companies, especially those involving interactive and active communication with mothers and pregnant women (eg, (2) online engagement with CMF companies, (3) CMF promotions and discounts), were associated with lower likelihood of optimal breastfeeding practices, which were partly influenced by positive maternal attitude towards formula. These results suggest that the Chinese governments and relevant entities should develop stricter legislation, regulatory and enforcement frameworks in line with the Code to comprehensively regulate the marketing practices of CMF companies, particularly related to informational, educational and promotional strategies and materials disseminated to the public and sale of formula products, thus protecting, promoting and supporting optimal breast feeding.