Abstract
The purpose of this study was to describe social norms and salient social identities related to breastfeeding intentions among African American mothers in Washington, D.C. Five focus groups were held with 30 mothers who gave birth to a child between 2016 and 2019. Two coders conducted pragmatic thematic analysis. This study demonstrated that women hold different identities relevant to making infant feeding decisions, with mother being primary and race/ethnicity, age, and relationship status factoring into how they define themselves. Mothers drew their perceptions of what is common and accepted from family, friends, the “Black community,” and what they perceived visually in their geographic area and heard from their health care providers. Mothers believed breastfeeding to be increasing in popularity and acceptability in African American communities in Washington, D.C., but not yet the most common or accepted mode of feeding, with some variability by socioeconomic status group. Implications for public health communication and social marketing are discussed.
Breastfeeding is a health promoting and risk reducing behavior for infant and mother. African American 1 babies’ rates of breastfeeding initiation, duration at 3 and 6 months, and exclusivity at 3 months improved by more than 10 percentage points each between 2010 and 2017 (Centers for Disease Control and Prevention, 2020). However, National Immunization Survey data show that Healthy People 2020 goals for breastfeeding have been achieved for White non-Hispanic babies born in 2017 but not for Black non-Hispanic babies (Table 1). Disparate breastfeeding rates are an important influence on African American health disparities over the life course (Spencer & Grassley, 2013).
Healthy People 2020 Targets and Breastfeeding Rates Nationally and in Washington, D.C.
CDC National Immunization Survey (NIS), among infants born in 2017 (Centers for Disease Control and Prevention, 2020). bCDC NIS 2011–2015, among infants born between 2010 and 2013 (most recent stratified analysis by geography and race/ethnicity; Anstey et al. 2017).
Factors influencing breastfeeding are summarized in Figure 1 within a social–ecological model. Bentley et al. (2003) posited that the complexity of breastfeeding disparities in African American communities stems from interactions between many influences across social–ecological levels. Social norms are one such influence that span levels, linking individual perceptions, attitudes, and intentions with interpersonal, organizational, community, cultural, and societal factors to influence behaviors. Empirical and qualitative evidence support that social norms are highly relevant in the breastfeeding context (e.g., DeVane-Johnson et al., 2017; Guo et al., 2016).
Social–ecological model of factors influencing breastfeeding in African American communities.
Theoretical Framework
This research followed Chung and Rimal’s (2016) revised framework of normative influences. “Norms are socially negotiated and contextually dependent modes of conduct” (Rimal & Lapinski, 2015, p. 394). There are multiple types of social norms and they are thought to function differently to influence health. Descriptive norms represent individual perceptions of what is commonly done within a group and injunctive norms are individual perceptions of what should be done (i.e., social approval; Cialdini et al., 1990). Subjective norms, related to but distinct from injunctive norms (Park & Smith, 2007), include perceived external expectations from important others and an individual’s motivation to comply with those expectations (Ajzen & Fishbein, 1970).
Per Chung and Rimal (2016), descriptive, injunctive, and subjective norms predict intentions and there are various attributes of the behavior (e.g., privacy), the individual (e.g., identity), and the context (e.g., injunctive norms) that moderate that relationship. To advance understanding of identity-related moderators of normative influences, ethnic pride, religiosity, collectivism, and temporal orientation were explored as possible facets of identity related to breastfeeding. These concepts were drawn from prior research on health and culture conducted with urban African American women (Kreuter et al., 2003; Lukwago et al., 2001).
Research questions that guided data collection and analysis were as follows:
Method
Participant Recruitment
This study was approved by the George Washington University Institutional Review Board on June 13, 2019 (NCR #191050). From June through August 2019, convenience sampling was used to recruit 36 mothers who self-identified as African American or Black (non-Hispanic), lived in Washington, D.C., and who gave birth to at least one child from 2016 to 2019. The sample size was expected to be sufficient to approach saturation of meaningful information related to the research questions (Carlsen & Glenton, 2011; Guest et al., 2017).
A socioeconomically diverse sample was achieved by promoting the study through partner organizations, including the D.C. Breastfeeding Coalition; obstetric, birthing, and pediatric care facilities; independent perinatal support providers; D.C. Special Supplemental Nutrition Program for Women, Infants, and Children (WIC); and social networks including African American affinity groups (e.g., local Mocha Moms chapter), neighborhood listservs, and a local Instagram account (@BreastfeedingInColor). Partners were provided with institutional review board–approved recruitment materials, which were further disseminated by others organically. A $100 gift card compensated participants for up to 2 hours and offset any costs incurred traveling to the focus group. All participants signed an informed consent document.
Data Collection
During telephone enrollment and after obtaining verbal consent, participant demographic information was entered into a REDCap (Research Electronic Data Capture) database via an interviewer administered survey. Five to 10 participants were enrolled per group and three to eight participated, leading to an analytic sample size of 30 due to no-shows. Five focus groups were conducted, each lasting 90 minutes. Three groups were held in community partner sites and two on the university campus. The second author, who has extensive training and experience in qualitative research in academic and private sectors, led focus groups according to a semistructured guide aligned with the theoretical framework (Table 2). The first author developed the guide and coauthors provided expert review. The facilitator was of the same ethnicity and gender as participants. The first author served as note taker. Participants were invited to use a pseudonym and were asked to respect the privacy of other participants and not discuss comments outside of the group. Focus groups were audio-recorded and transcribed by a professional transcriptionist.
Theoretical Alignment of Focus Group Questions.
Data Analysis
The first and second authors 2 conducted iterative, pragmatic thematic analysis (Burnard, 2011) using MAXQDA Plus 2018 software (VERBI Software, 2018). Initial codes were developed based on the theoretical framework and codes added as needed. Coders independently applied codes to transcripts. Multiple discussions of interpretations and revisions to the coding scheme showed strong agreement. Coders conducted final readings of all five transcripts applying the final codebook. They held additional meetings where coding patterns were used to identify themes and to discuss the organization and presentation of findings. Participants received a two-page summary of the results via email, requesting general feedback and specific critiques on the identity findings. Only two participants responded, affirming the findings and reiterating the importance of economic and workplace factors in breastfeeding outcomes.
To approximate social position, participants were categorized into two socioeconomic status (SES) groups. Washington, D.C., is divided into eight wards. Participants who resided within either Ward 7 or 8 in the D.C. were initially categorized as “lower SES.” Household income was not directly collected; however, according to the American Community Survey data, median household income in those wards was $37,736 between 2014 and 2018 compared with $94,571 across Wards 1, 2, 4, 5, and 6, which were represented in the sample (DC State Data Center, n.d.). Unemployment rates from 2014 and 2018 averaged 16.8% in Wards 7 and 8 compared with 5.6% in other wards represented in this sample. Educational attainment and subjective self-reported financial status were then reviewed causing three participants who reported being homeowners in Ward 7 with master’s or doctoral degrees and who reported “living comfortably” to be recategorized as “middle to higher SES.” One participant with a master’s degree was intentionally left in the lower SES group because she lived in Ward 7 and reported not being able to meet basic expenses. Six participants who reported not (1) or just meeting (5) basic expenses but who had achieved post–high school education and lived in wards with higher median income and employment statistics were categorized as middle to higher SES. Coded content was then compared by SES group.
Results
Table 3 illustrates the sample characteristics overall and by SES group. Most participants listed a history of breastfeeding even if only a brief duration or with prior children even if currently using formula. Formula supplementation while breastfeeding was common. Table 4 provides a summary of key findings.
Demographic Characteristics of Participants.
Note. SES = socioeconomic status.
Summary of Key Findings by Research Question.
Note. SES = socioeconomic status; WIC = Special Supplemental Nutrition Program for Women, Infants, and Children.
Normative Referents Varied
The most commonly invoked normative referents (based on code frequency) included immediate family members and close friends, who were also described by participants as the most important to their infant feeding decisions. Participants specifically mentioned the influence their own mothers had and other female relatives, particularly aunts. The next most common reference group was Black communities (e.g., “Black people,” “Black folk/s,” “Black community,” “Black doctors,” “Black mothers/moms,” “Black women”) followed by geographic community. Health care providers were commonly mentioned and described as important to decision making. Partners/infants’ fathers were occasionally referenced but were not reported to have a large role in the decision-making process.
Perceptions of Breastfeeding Social Norms
Top-level themes included a uniform perception that descriptive and injunctive norms have shifted in favor of breastfeeding in recent years (trending norms) and variation existed in perceived norms by the referent and SES groups.
Descriptive Norms
Overall participants agreed that breastfeeding is not common or “popular” in African American communities in Washington, D.C. However, there was extensive discussion about how the prevalence has shifted over time such that breastfeeding is becoming more common. This was described by mothers who were comparing community norms today with those of their parents’ where formula was most prevalent. Mothers with older children also perceived norms shifting in favor of breastfeeding within a more recent time period between their first child and most recent baby. The sense that breastfeeding was trending upward appeared to apply mainly to initiation; perceptions of norms around duration and exclusivity varied among this sample of mothers.
I think now like she said there’s a trend, there’s a big trend and they’re pushing it but I don’t think before they were . . . So now there’s a different trend and you see a shift in breastfeeding being a new thing. (Lower SES, age 37 years)
SES Differences
Across SES groups, several mothers noted being the first in their families to breastfeed but perceptions of descriptive norms differed somewhat by SES group. Middle to higher SES mothers perceived breastfeeding as uncommon in the broader African American community, particularly among younger mothers. These mothers also referred to historical uncommonness of breastfeeding in their own families but noted that among their current friend and social circles that breastfeeding and breastfeeding for longer durations (e.g., 6 months minimum) was the norm, attributed to increased resources and support. Middle to higher SES mothers also shared examples of opportunities they had to see breastfeeding in public, which they described as shaping their perception of normality, noting that they were often in communities that were not exclusively African American: Because the things that I go to I’m usually the only Black woman that’s there . . . And so that’s just like it was so normal that I felt like okay this is fine I can do it anywhere, like who cares. (Higher SES, age 36 years)
Whereas another participant stated, “I’ve lived here forever and it’s just like this is a different like, D.C.’s a whole different culture now. So like it’s awesome but when I was growing up I’ve never seen anybody breastfeed” (Lower SES, age 30 years). For lower SES mothers, there were mixed perceptions on how common breastfeeding is in African American communities. Even among those who breastfed their child(ren), several did not know many other breastfeeding mothers personally, while several others mentioned having close friends who breastfed.
One of my girlfriends, the godmother of my son is the only friend that I have that breastfeeds. And I joined a community on Facebook; they said Black women do breastfeed but I don’t know too many. My sister breastfed for three months but when she went back to work it was like a lot of jobs don’t offer you the space to pump and her supply kind of dried up. But I didn’t know too many people. And my mom she definitely didn’t breastfeed. (Lower SES, age 30 years) I actually see a lot more young mothers breastfeeding. I have a whole lot of friends that breastfed their babies. It’s become more common. (Lower SES, age 24 years)
Injunctive and Subjective Norms
Injunctive norms
Results on whether breastfeeding behaviors are approved of in African American communities in D.C. were context or referent dependent. There was general agreement that breastfeeding is becoming more accepted in African American communities but that there are generational differences in approval. Participants shared stories of disapproval from older family or community members from an older generation and attributed these differing views to education and awareness that were inhibited by targeted formula marketing and promotion of breast milk supplements by health care providers and the government in decades past. There was substantial discussion indicating disapproval related to stigma, fear, and shaming for nursing in public. For example a 23-year-old participant in the lower SES group said, “I feel shamed when I breastfeed. I think it’s generally disapproved. I think that no matter where I’ve gone, cover-up, no cover-up, people look at me like something’s wrong when I’m breastfeeding.” Negative perceived injunctive norms were also apparent in stories about nursing at home with others present, especially men. Mothers alluded to concerns about modesty and what is or is not appropriate when nursing in the presence of others, but there was disagreement about whether mothers should have to cover up. Participants’ language also reflected the dynamic between private and public feeding and suggested negative injunctive norms. Words such as “whip out,” “slap out,” and “titty” tended to be used to talk about feeding in front of other people. These words were taken to imply that breastfeeding was immodest and unacceptable, especially in front of others. Words such as “feed,” “nurse,” “boob,” and “breast” were most often used to talk about feeding in private and conveyed a positive or neutral meaning.
Both mother’s age and child’s age were relevant to perceptions of approval of breastfeeding. Two participants in separate focus groups suggested that it was not appropriate for young, teenage mothers to breastfeed because of the stigma they already face, the sexualization of breasts, as well as the incompatibility of breastfeeding with staying in school. There was also discussion about increasing disapproval as the child aged, though no clear agreement on when a child should be weaned. Most mothers perceived it was acceptable to continue nursing for the child’s first year, which aligns with the recommendation of the American Academy of Pediatrics (2012). Beyond that initial year as an infant, perceived disapproval increased as the child aged, even when the child was within the World Health Organization (2011) recommended age for continued breastfeeding up to 2 years. There was agreement that breast milk continued to confer benefits, but feeding at the breast was unacceptable as the child’s teeth erupted and the child became more independent. Several mothers shared similar stories of feeling pressure from family members to introduce water or solid foods before they were comfortable or felt the baby was ready (and before the American Academy of Pediatrics guidelines recommend).
Subjective norms
Findings were also mixed on perceptions of expectations around infant feeding. Influence and expectations from family, close friends, and healthcare providers (Davis et al., 2021) were common but highly variable as to whether they encouraged breastfeeding or formula/bottle-feeding. Reasons for the pressure also varied from cost-effectiveness, to pressure to return to work, to health benefits, to implied skepticism that the mother could succeed at breastfeeding.
A lot of young Black moms don’t breastfeed, they bottle feed. And again, my sister says “well she’s been feeding now, she’s gained a lot of weight. Why don’t you put her on the bottle?” No my breasts are working fine. I’m not going back to work. I like the bonding time.” (Higher SES, age 38 years)
I was very offended and I’m at the hospital and she’s like ‘well breastfeeding doesn’t work for everybody so let me give you this just in case.’ But I was like irritated that, and it was like somebody could definitely use this stuff, but it’s like if I’m telling you that I’m breastfeeding why wouldn’t you send the lactation specialist in here for a good pep talk before I leave instead of giving me this bag of stuff that I have no intentions on using? (Lower SES, age 27 years)
SES differences
There were limited clear differences in perceived injunctive and subjective norms by SES group. However, there was a shared belief among lower SES mothers that changes in policies and messaging from WIC in the last 10–15 years were a driving force in shifting attitudes and norms toward increasing commonality and acceptance. One 37-year old mother said, “the government is now starting to promote breastfeeding but before they wasn’t, they was promoting you getting that formula” and a 33-year old mother said about WIC’s enhanced nutritional package, “It just seems so much easier because now in Ward 7 and 8 I’ve noticed they make it easier for you to go breastfeed.”
Salient Social Identities
Across the sample, the most salient identity was that of mother, with roles depicted as caregiver and protector, implying that any infant feeding decisions were made with the best intentions. Variants on this identity emerged and included self-described “Black moms,” “young moms,” “single moms,” and “mature” or “seasoned moms.” These identities were not mutually exclusive and were held by some simultaneously and by others at different stages of motherhood.
Comments from Black mothers reflected recognition of the intersectional stress and grind from being African American, a woman, and a mother. They exhibited pride in their resilience in the face of historical and current discrimination and economic challenges and many sought to build community with other African American women. Young mothers were characterized as having had their first babies in their teens and participants described themselves or other mothers they categorized this way as lacking awareness, self-confidence nursing in public, “patience,” time, independence, or support for breastfeeding. Single mothers typically made decisions about infant feeding based on their finances and work situations and described having less support than mothers with partners had. Mature/seasoned mothers reported that age conferred self-confidence so the opinions of others did not matter to them, and those newly entering motherhood later than average were determined to do things “right” by breastfeeding.
Ethnic Pride
For some participants, identifying as a Black mother led them to defiantly choose breastfeeding to counter poor health outcomes and actively work to shift norms through their own behavior, exhibiting ethnic pride: I purposely breastfeed in public so that people can see that I’m a Black mother who breastfeeds [so] it becomes normal, to make a new normal . . . I feel like I want to give my Black daughter the best opportunity that she can have to be an awesome Black woman. (Higher SES, age 38 years).
Examples of intentionally working to shift norms included nursing in public, including uncovered, offering verbal encouragement to other African American mothers nursing in public, and promoting breastfeeding to other African American women in their sphere (Davis et al., 2021).
Religiosity, Temporal Orientation, and Collectivism
Neither religiosity, temporal orientation, nor collectivism were reported by this sample as accessible or salient components of their social identity in decision making around infant feeding. However, the mothers’ life circumstances, including available social supports, were immensely important to their decisions and they chose whichever infant feeding method was most feasible.
Discussion
Normative Referents Drive Normative Perceptions
Consistent with prior research, this study demonstrated that norms were variable by referent. While family and close friends were referred to most often and cited as influential to infant feeding decisions, gender and age of family and friends often dictated whether they approved of breastfeeding or not. Perceptions of behavioral prevalence also varied by referent age, with mothers deriving their perceptions mainly from women in their family and women similar to them in age, ethnicity, and values. These results are similar to findings by Lapinski et al. (2015), who found that descriptive and injunctive prenatal care norms in rural Mexico were referent dependent.
Norms Consistent With Trends Indicate Positive Deviance
Participants’ perceptions that breastfeeding is not common in African American communities may reflect awareness of documented disparities or be inferred from exposure to local and national breastfeeding promotion efforts targeting African American communities such as Black Breastfeeding Week (blackbreastfeedingweek.org), WIC marketing, and campaigns like It’s Only Natural (womenshealth.gov/its-only-natural). Perceptions that breastfeeding is becoming more common and accepted are consistent with trend data (Anstey et al., 2017). Positive breastfeeding behavior is typically associated with higher levels of income and education (Anstey et al., 2017; McKinney et al., 2016). In this study, mixed results around family members encouraging or discouraging breastfeeding regardless of SES may be due to some middle to higher SES women having close family contacts who are lower SES (Pattillo, 2013). The perception of lower SES women of the role of WIC in changing norms aligns with the 2004 reinvigoration of the Loving Support social marketing and peer counseling program and the 2009 policy changes to increase nutrition benefits for breastfeeding mothers (Collins et al., 2010; Whaley et al., 2012). Perceptions of injunctive norms were most often described relative to breastfeeding in public or in front of others, likely because interpersonal communication plays a large role in diffusion of normative information (Lapinski & Rimal, 2005).
Many participants believed that breastfeeding was neither common nor socially accepted, yet, almost all of the participants had breastfeeding experience. Indeed, some language reflected internalized injunctive norms in favor of breastfeeding; phrases such as “push through,” “sheer determination,” “soldier point,” and “persevere” indicated an intense desire to successfully breastfeed despite often significant challenges. In this instance, the traditional social norms approach to correct misperceived norms to influence behavior would be inappropriate. Cialdini (1987) suggests that perceived descriptive norms serve as social proof that the prevalent group behavior is the correct behavior. Furthermore, conformity with perceived injunctive norms demonstrates an individual’s motivation for unity with the group (Bendor & Swistak, 2001). The fact that most women recruited had some breastfeeding history despite perceiving weak descriptive and injunctive norms in African American communities may reflect positive deviance, as others have suggested before (Barbosa et al., 2017; Gross et al., 2017; Ma & Magnus, 2012). According to Spreitzer and Sonenshein (2004), “positive deviants” are individuals who choose to go against a group norm with honorable intent, such as to enact a prosocial or healthy behavior; in other words, they are community role models. One participant described her decision as such: It was a lot of people that didn’t know a lot about breastfeeding so when I made that decision it was mostly frowned upon. Friends being like “you going to do that?” And it was very discouraging but I still breastfed.
Social Identities Were Important in Forming Intentions
Different identities were relevant when making decisions around infant feeding. The variations on mother were not mutually exclusive, but held differing connotations for what assets and challenges influenced the decision to breastfeed or not, and for how long. Ethnic pride appears to be a promising construct to consider in breastfeeding interventions. While most participants did not perceive their race had any role in their infant feeding decisions, several mothers shared reasons for breastfeeding that implied pride in their ethnic group. This positive association is consistent with research on African American pride in other health contexts (Belgrave & Abrams, 2016; Kreuter & Haughton, 2006) and in breastfeeding for Hispanic mothers (McKee et al., 2004). Ethnic pride may be related to positive deviance, empowerment, and breastfeeding activism, consistent with what others have found (Spencer et al., 2015). Future research should explore the potential for helping to foster ethnic pride and for activating pride in breastfeeding messaging and interventions.
Participants did not describe a connection between religion or religiosity and infant feeding choice, yet about half reported being religiously affiliated or identifying as spiritual, humanist, or universalist. These findings differ somewhat from past research that found a prominent role of spirituality in supporting a decision to breastfeed (Spencer et al., 2015). This may be a reflection of the younger age of the sample and a movement away from organized religion among African American millennials (Jordan, 2019; Pew Research Center, 2014). The tendency among millennials to identify as more spiritual than religious makes spirituality an important factor to explore further in targeting messages and promotion channels.
Strengths and Limitations
This study had strengths including the diverse sample, which resulted in varied perspectives from mothers across SES groups and the dual-coder approach to data analysis. Geographic restriction to Washington, D.C., means findings may not be wholly transferable to other predominantly African American communities. Limitations include use of self-nominated volunteer participants who may not fully represent the broader population, and inherent challenges maintaining focused discussion and balanced participation. Using a subjective measure of financial status instead of directly asking household income was a limitation; participants may have interpreted response options like living comfortably or not meeting basic expenses differently. This may have affected our SES categorization; however, we do not believe the findings would have been greatly changed by this. Focus group size also varied, which allowed more speaking time for individuals in smaller groups and less time per person in larger groups, potentially skewing findings. Participation by a very small number of mothers who never breastfed prohibited comparison of their perceptions of norms to mothers who breastfed. For study promotion, we worked with local birthing facilities, breastfeeding support professionals, and social service organizations in an attempt to reach mothers with and without breastfeeding experience. The high number of participants with breastfeeding experience might be due to more enthusiastic promotion among breastfeeding advocates or may simply reflect increased breastfeeding initiation tends in the city.
Implications for Theory and Practice
To eliminate breastfeeding disparities, a coordinated national social marketing approach targeting all levels of the social–ecological model is needed (Pérez-Escamilla & Chapman, 2012). Mass and social media have the potential to serve as critical channels for social marketing around infant feeding (Asiodu et al., 2015; Bylaska-Davies, 2015). Understanding social norms is necessary if a communication intervention is being considered and offers a useful framework for the design of social marketing campaigns (Mabry & Mackert, 2014).
Based on these study results, public health practitioners should aim to strategically incorporate social identity and messages should be used to increase the salience of referents emanating desirable norms in specific contexts (Larimer et al., 2009). Ethnic pride may be relevant for audience segmentation. Future social marketing of breastfeeding toward African American audiences could appeal to positive deviance through messages promoting breastfeeding in communities where rates are low (Singhal, 2010). One approach could be to follow recommendations of Sparkman and Walton (2017), who proposed using trending norms messages to encourage sustainable behaviors. Their suggestion was to develop messages that highlight the increasing prevalence or popularity of a behavior, and their field experiments found trending norms messages to be more effective than traditional static descriptive norms messages at predicting intentions. Leveraging prototypical in-group community members who have successfully breastfed to deliver health promotion messages and share strategies for overcoming challenges (Hogg & Reid, 2006) could increase visibility of Black breastfeeding and further enhance message effectiveness. Given that WIC was mentioned as a facilitator of the shifting norms and source of breastfeeding support, efforts should be expanded in Washington, D.C., to ensure all eligible women are enrolled in the program (Trippe et al., 2019).
Footnotes
Acknowledgements
The authors would like to acknowledge the participants who generously shared their time and entrusted their stories to us. The authors also thank their community partners who facilitated recruitment, and Mamatoto Village and the East of the River Lactation Support Center for serving as focus group sites. The authors also acknowledge the advice of coinvestigator, Anayah Sangodele-Ayoka, CNM, MSN, MSEd, Clinical Instructor, George Washington University School of Medicine and Health Sciences.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received financial support for the research, authorship, and/or publication of this article: This research was generously supported by the Clara Schiffer Fellowship for Women’s Health and a fellowship from the Sumner M. Redstone Global Center for Prevention and Wellness. REDCap infrastructure that made this project possible was partially supported by the National Institutes of Health National Center for Advancing Translational Sciences (Award No. UL1TR001876). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Advancing Translational Sciences or the National Institutes of Health.
