Breastfeeding and Intersectionality in the Deep South: Race, Class, Gender and Community Context in Coastal Mississippi

Abstract

Intersectionality, especially with a race–class–gender focus, has been used to study many facets of women’s experiences. However, this framework has been underutilized in the study of breastfeeding prevalence. Our study is the first of its kind to use intersectionality to illuminate breastfeeding network prevalence disparities with empirical data. We use insights from this theory to examine breastfeeding patterns reported by women living on the Mississippi Gulf Coast. Mississippi makes an excellent site for such an examination, given its history of racial discrimination, entrenched poverty, and strikingly low rates of breastfeeding, particularly for African American women. We identify a series of factors that influence racial disparities in lactation network prevalence, that is, breastfeeding among friends and family of the women we surveyed. Our investigation relies on survey data drawn from a random sample of adult women who are representative of the Mississippi Gulf Coast population supplemented by a non-random oversample of African American women in this predominantly rural tri-county area. Results from the first wave of the CDC-funded 2019 Mississippi REACH Social Climate Survey reveal that Black-White differentials in breastfeeding network prevalence are significantly reduced for African American women who report (1) higher income levels and (2) more robust community support for breastfeeding. We conclude that breastfeeding is subject to two key structural factors: economic standing and community context. An appreciation of these intersecting influences on breastfeeding and long-term efforts to alter them could bring about greater breastfeeding parity among African American and White women in Mississippi and perhaps elsewhere. We end by identifying the practical implications of our findings and promising directions for future research.

1. Introduction

Mississippi has long been a leading U.S. state in poor health outcomes, often due to a variety of social determinants across race, class, and gender [1]. In 2023, Mississippi’s Black population, especially those reporting under USD 35,000 in annual household income, was the most uninsured while also having the highest prevalence of heart disease, hypertension, stroke, diabetes, renal disease, COVID-19, AIDS, various forms of cancer, septicemia, and homicide, along with mortality linked to these conditions [2]. Overall mortality rates for Black Americans were found to be higher across Gulf Coast states, including Mississippi, in comparison to White Americans across the United States and Black Americans in other states [3]. Perinatal statistics for Mississippi indicate a preterm birthrate of 15.0%, nearly five points higher than the U.S. national average (10.4%) [4]. However, the preterm birthrate for Black Mississippi infants was 1.4 times higher (18.2%) than those for any other racial or ethnic group.

The chronic health conditions that adversely affect Black Mississippians, especially along the Gulf Coastline (i.e., smoking, hypertension, unhealthy weight, diabetes) increase the likelihood of a preterm birth [4]. The Mississippi infant mortality rate (9.1) is almost 4 points higher than the national average (5.6), with Black infants dying at a rate 1.3 times higher than the state’s average. Similarly, Mississippi’s maternal mortality rate, or the death rate of birthing women from pregnancy or post-childbirth complications, is 39.1 versus the 23.2 national average [4]. Breastfeeding acts as a protective factor against a myriad of diseases and cancers that disproportionately affect Black Americans, especially those in Mississippi. Human lactation reduces the development of acute and chronic diseases, which are the strongest health risk indicators for Mississippi women and, specifically, mothers [5,6]. Across the U.S., 8.6% of live births met standards of low birth weight in 2022, with 86 infants born per week at low birth weights across Mississippi [7]. Breastfeeding is the most highly recommended practice for infant nutrition by leading pediatricians in the World Health Organization (WHO) and the American Academy of Pediatrics, with infant nutrition experts strongly emphasizing this practice for populations at greatest health risks (low-income, racial disparities, etc.) [8,9,10].

The theory of intersectionality was originally introduced by scholars who focus on women of color. Intersectionality is designed to be inclusive by focusing on the linkages among various social identities that can collectively reinforce marginalization [11,12,13]. Intersectionality as an analytic framework for examining sociopolitical issues that emerge from race, gender, class, and other socially defined categories is especially relevant in breastfeeding [14]. The use of intersectionality in examining breastfeeding social networks introduces a deeper understanding of the interconnecting social and structural components related to breastfeeding differentials. By developing a greater knowledge of the components that interact with and influence breastfeeding, promising pathways for addressing gaps in care and nutrition can be established.

Several leading factors related to breastfeeding prevalence in social networks are prominently communicated through an intersectional lens: race/ethnicity, economic standing (income), and community context (lactation networks, spaces, messaging, etc.) [15,16,17]. First, why is breastfeeding of social importance? Public health benefits of breastfeeding include reductions in infant adverse health outcomes (i.e., respiratory infection, diarrhea, and infant mortality), especially throughout low-income communities. Breastfeeding also positively affects neurodevelopment, which reduces the risk of developing significant mental health disorders. Second, Black women’s breastfeeding rates are among the lowest of all ethnic groups in the United States, but they are also generally knowledgeable about breastfeeding benefits for infants. Unfortunately, fear of pain, lack of time allotted to breastfeeding, or unaddressed issues (i.e., latching, milk production) are primary barriers to breastfeeding for low-income Black women [18,19,20,21,22]. Involvement in community organizations and exposure to communications addressing questions, concerns, and misinformation are key aspects of intersectionality in breastfeeding that directly impact breastfeeding initiation and duration.

The next element of intersectionality includes community factors of breastfeeding that are influential at several levels (i.e., individual, interpersonal, organizational, and policy). Individually, breastfeeding is a known, valuable practice. However, women of color are more likely to encounter extreme exhaustion, isolation, and challenges associated with the time required to breastfeed, and are more inclined to lack self-efficacy or confidence in their breastfeeding abilities [23,24]. When seeking breastfeeding support or connection, in-person and virtual communities (breastfeeding support meetups, listservs, group chats, social media pages, etc.) are also viable pathways to knowledge. Through these groups, mothers interact with experts such as lactation specialists, other healthcare professionals, and fellow breastfeeding mothers on varying topics (pumping, excretion, storing milk, etc.) [25,26,27,28,29,30,31,32]. Research has previously documented that social media and mainstream media marketing are prevalent among Black women in Mississippi. However, there remain low levels of reported breastfeeding compared to White Mississippians [33]. Interpersonal relationships, cultural traditions, media platforms, peer-to-peer learning, and family experiences can act as supports or barriers for breastfeeding, depending on the informational content about breastfeeding and the durability of exposure to specific perspectives on this topic. Organizationally, healthcare centers such as hospitals often lack personnel with breastfeeding experience or lack breastfeeding support policies altogether [34]. Interventions at the interpersonal and policy levels delivered by Women and Infant Children (WIC) and other healthcare agencies have improved breastfeeding outcomes among women of color [35]. Additionally, educating family members (especially fathers), community members, and healthcare staff has been a successful method of increasing breastfeeding dialogue and intent while reducing maternal fear about breastfeeding [36,37,38,39].

While there has been extensive research into breastfeeding practices and relationships, our study of intersectionality in the context of Mississippi breastfeeding social connections is unique. One similar intervention was conducted in rural Mississippi, focusing primarily on race relations in breastfeeding. Delta Healthy Sprouts enlisted the involvement of African American female participants (79 out of 82; 96.3%) [40]. A study of Delta Healthy Sprouts revealed that this intervention increased breastfeeding knowledge from baseline to the late gestational stage. Quite notably, there was also an increase in social breastfeeding beliefs exhibited by mothers who initiated breastfeeding. The Delta Healthy Sprouts intervention demonstrated the necessity of increasing breastfeeding knowledge while addressing modifiable societal norms and social support. Prior research has not prioritized a lens of intersectionality related to breastfeeding in Mississippi or, more broadly, the United States. Our current study focuses on the intersectional nature of breastfeeding and social network embeddedness, especially related to income and race/ethnicity in Mississippi. Our present study examines African American and White women’s respective social network differentials in lactation prevalence, as associated with (1) reported income levels (social class) and (2) perceived community support for breastfeeding (beliefs about the collective acceptance of breastfeeding). Based on the previous scholarly literature (reviewed above), we hypothesize that breastfeeding network prevalence will be positively associated with household income (H1) and higher levels of perceived community support for breastfeeding (H2) among African American women. We do not argue that these factors are ineffectual for White women but anticipate that they will be more robustly influential for African American women. The results of our study offer key insights for effectively addressing persistent barriers to breastfeeding in Mississippi and across areas with similar racial/ethnic demographics and health disparities.

2. Results

Table 1. Sample characteristics for Black and White women (unweighted n = 241).

	n	Percent	Mean	SD
Breastfeeding Network Prevalence	-	-	2.48	1.58
African American	120	49.79		-
White	121	50.21	-	-
Household income in 2018	-	-	5.34	3.07
Community Support (index)	-	-	3.26	0.99
Age	-	-	43.66	17.00
Education	-	-	5.07	1.53
Employed	129	53.53	-	-
Not working/Other	112	46.47	-	-

presents the descriptive statistics for women participants of the Mississippi REACH Social Climate Survey. This study exclusively included White and African American women, with nearly equal representation: White women constituted 50.21% of the sample, while African American women made up 49.79%. At the time of survey completion, 53.53% of the women were employed. The average age of the female respondents was 43 years, and their average educational attainment was at the level of some college or vocational school.

Interpretations of Regression Results

We estimate four nested ordinary least squares (OLS) regression models to test our research hypotheses. As detailed in

Table 2. OLS regression models to predict breastfeeding prevalence in primary social networks among Black and White Women (weighted n = 241).

Variable	Model 1		Model 2		Model 3		Model 4
African American (White = reference)	−1.216	***	−1.211	***	−1.049	***	−0.994	***
Household income	0.129	*	0.020		0.085	*	0.080
Household income x AA			0.159	**
Community support (index)					0.264	**	−0.174
Community support x AA							0.659	**
Age	−0.006		−0.006		−0.008		−0.010
Education	0.076		0.081		0.112		0.139
Employed (other = reference)	−0.118		−0.049		−0.055		0.024
Intercept	2.371	***	2.940	***	1.512	*	2.930	***
F	16.911	***	15.649	***	12.166	***	12.585	***
R2	0.265		0.286		0.235		0.271

Note: * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001.

, Model 1 incorporates race and household income, along with statistical controls to assess the regression relationship between household income and breastfeeding network prevalence. Model 2 examines the intersection of race and household income. Model 3 investigates whether there is a positive association between perceived community-based peer support for breastfeeding and lactation network prevalence, after accounting for race and other sociodemographic characteristics. Finally, Model 4 explores the interaction effect of race and community-based peer support for breastfeeding on lactation network prevalence.

In reporting our results, we refer to Table 2 and Figure 1 and Figure 2. This table and these figures allow us to interpret our findings in relation to our hypotheses. Table 2 displays the OLS regression coefficients along with levels of statistical significance (p-values). Interaction effects are also statistically presented in Table 2 (Models 2 and 4). Figure 1 and Figure 2 visually depict interaction effects among key variables using line graphs. As anticipated, Model 1 in Table 2 indicates that African American women in our sample report significantly lower levels of breastfeeding network prevalence than their White counterparts net of statistical controls (β = −1.216 and p < 0.001). Additionally, the regression coefficient for household income is positive (β = 0.129) and statistically significant at the 0.05 level. Based on these results, Model 2 in Table 2 and Figure 1 further explore the interaction effect of race and income. Figure 1 provides compelling visual evidence of the intersection of race and household income in relation to breastfeeding network prevalence. Specifically, the predicted value of breastfeeding network prevalence increases with each unit increase in household income, controlling for race and other sociodemographic characteristics. However, the effect of increased household income on breastfeeding network prevalence is more pronounced for African American women than for White women on the Mississippi Gulf Coast. This contrast is evident through the distinct slopes shown in Figure 1, along with the statistically significant coefficient in Model 2 (β = 0.159 and p < 0.01).

Table 2 also features two additional models (Models 3 and 4) that examine the associations between race, community support for breastfeeding, and breastfeeding network prevalence. The negative association for African American women and the positive association for household income in Model 3 largely mirror those presented in Model 1. However, the community support for breastfeeding variable in Model 3 is significantly and positively associated with lactation network prevalence (β = 0.264 and p < 0.01). Additionally, Model 4 and Figure 2 further test the interaction effect of race and community support on breastfeeding network prevalence. As displayed in the table, the interaction effect is statistically significant (β = 0.659 and p < 0.01), which is visually illustrated in Figure 2. As depicted in Figure 2, there is a racial crossover effect, where community support for breastfeeding predicts lactation network prevalence differently for African American and White women. Specifically, the regression slope for community support is positive for African American women but negative for White women. These regression results collectively support our intersectionality hypotheses involving gender, race, economic status (social class), and community support in the context of breastfeeding. In summary, Black women in higher-income homes and those who perceive greater community support for lactation report more robust breastfeeding network prevalence, comparable to their White peers. Constrained income and low community support significantly inhibit breastfeeding network prevalence among Black women.

3. Discussion

Breastfeeding is a significant aspect of the mother–infant relationship and, to a wider degree, the connections between mothers and the community in which they live. In the United States, Mississippi has some of the highest preterm birthrates, frequent low birth weights, and increased health risks for African Americans and women of color [2,3,4,5,6,7]. Breastfeeding is the most recommended infant nutrition source for positive child and adult health outcomes while also being a practice dependent on a myriad of interconnected social factors [8,9,10]. Intersectionality has its foundation in women of color research and should be a significant aspect of many breastfeeding initiatives, as women of color are often disproportionately affected by low breastfeeding rates and adverse health outcomes [17,18,19,20,21,22,23,24]. For the most part, intersectionality has not been explored in the context of breastfeeding, and certainly not with attention to the factors we have explored here. Studies have highlighted the individual but socially diverse aspects of breastfeeding and the importance of identifying gaps in messaging and health care [23,24,33,34]. However, the intersectionality of breastfeeding is an understudied, interconnected relationship between gender, race, and social class. Our study was the first of its kind to use an intersectional lens in analyzing the convergence of breastfeeding peer support, breastfeeding network prevalence, women of Black and White racial groups, and household income in Mississippi. We hypothesized that the intersecting elements of race, income levels, and social support would be prominently evident in opposing dynamics among African American and White women. Confirming our intersectional hypotheses, our results indicate that race, social class (income), and community support are interrelated and have significant impacts on our outcome variable, namely, breastfeeding network prevalence.

Controlling for sociodemographic characteristics, African American women reported lower levels of breastfeeding network prevalence (i.e., the proportion of respondents’ family or friendship networks where breastfeeding occurs) compared to White women. However, this racial disparity in lactation network prevalence between African American and White women is significantly mitigated among African American women who report (1) higher income levels and (2) greater perceived community support for lactation (i.e., acceptance of breastfeeding in community spaces). Consequently, when African American women have household incomes comparable to those of White women and perceive their community as more supportive of breastfeeding, both groups exhibit similar levels of breastfeeding exposure within their primary social networks. These interaction effects depicted in Figure 1 revealed that although both African American and White women benefit from higher household income in terms of lactation network prevalence, African American women experienced a more pronounced benefit. Additionally, Figure 2 displays a positive relationship between community support and breastfeeding network prevalence for African American women, but this relationship is negative for White women. Such a racialized pattern may contribute to the narrowing of the gap in breastfeeding network prevalence between African American and White women in Mississippi.

Our results indicate that breastfeeding is subject to two interconnected structural influences, economic standing and community context, underscored by social factors of gender and race. We, therefore, surmise that utilizing an intersectional lens, the Black–White gap in lactation network prevalence is unlikely to be closed by a sole focus on household income or community support. The racial gap in lactation network prevalence is subject to various intersecting influences. Breastfeeding initiation and duration are impacted by economic standing (income/social status) due to the availability of transportation, healthcare access, breastfeeding education, peer support, and, overall, more varied modes of breastfeeding exposure [17]. Community context is also influential for potential breastfeeding mothers as it determines social support. For example, does the congregation she attends have breastfeeding-friendly spaces? Does her job have a lactation room? Are her family and friends educated or experienced in breastfeeding? This last point is vital as studies have shown the key role that fathers and supporting family members play in breastfeeding awareness [26,30,36,37,38].

Intersecting influences on breastfeeding and long-term efforts to alter them could bring about greater breastfeeding parity among African American and White women in Mississippi. Introducing policies for adoption in the healthcare system (i.e., hospitals, care centers, workplaces) that address breastfeeding support and social connectedness would be a significant step forward in closing racialized breastfeeding gaps. Expanding breastfeeding knowledge, experience, and acceptance across age groups, genders, and racial/ethnic groups should be a primary goal of long-term maternal healthcare policies and practices. A state-specific positive change could be replicated throughout other locations with similar racial, gender, and health disparities. African American women, along with other non-White racial and ethnic groups, continue to be at the highest risk for infant and mother health challenges that could otherwise be reduced by breastfeeding [5,15]. A prolonged focus on African American and other non-White breastfeeding practices and educational promotion is needed.

This study explored a relatively unique area of intersectionality within the field of breastfeeding, yet there were a few limitations that suggest avenues for future research. The analysis used data from a baseline community assessment of Mississippi Gulf Coast breastfeeding network disparities between African American and White racial groups. Our research team took every effort to examine the content validity of survey items specific to Gulf Coast Mississippians. However, future studies aiming to replicate or extend our results in other contexts would do well to cross-check content validity for those populations, particularly concerning breastfeeding network prevalence given the novel nature of this measure. It is worth noting that the survey items related to breastfeeding community support were adapted from a CDC survey with established construct validity. Future studies could build upon our methodologies and findings by incorporating item validity tests tailored to specific populations (i.e., race/ethnicity, age, gender/sex, education, religion). Comparative analyses of generalized versus situational aspects of community breastfeeding acceptance are also encouraged. Additionally, the initial survey (wave 1) was conducted before the COVID-19 pandemic and subsequent formula shortages. The second wave of the Social Climate Survey, completed in 2023, investigated the impact of the COVID-19 pandemic and formula shortage on community breastfeeding knowledge, dialogue, and experiences throughout the Mississippi Gulf Coast. Analyzing this second wave of data is crucial for understanding post-pandemic breastfeeding dynamics and informing actionable steps in the field.

4. Materials and Methods

The current study analyzes validated data that were utilized in two previously published articles on racialized breastfeeding media messaging differences [33,41]. The present study has a distinct focus separate from the previous works, which did not explore the relationships between race and household income in a community context for breastfeeding support. A cross-sectional survey design was utilized to identify the reported prevalence of breastfeeding in primary social networks, racial identity, and household income, among other factors. A cross-sectional design is appropriate for establishing associations among the variables of interest (independent and dependent variables) net of confounding factors (controls) but cannot establish causal order. This design was selected using the Mississippi REACH Social Climate Survey as a baseline assessment of a community-based project established in Mississippi from 2018 through 2023. Mississippi REACH prioritizes addressing health disparities adversely affecting African Americans across several Gulf Coast counties. Therefore, this study’s population sample adhered to the MS REACH counties served. This survey activity was designed to form a generalized understanding of African American and White social disparities in breastfeeding along the Mississippi Gulf Coast and similar areas. Data collection approval was determined by an existing memorandum of understanding between the Declaration of Helsinki and the Institutional Review Board of Mississippi State University (IRB-17-04-MOU). Data collection through survey administration was conducted from July through September 2019.

The analytical sample for this study came from a subsample of the full Mississippi REACH Social Climate Survey dataset. The analytical sample used in this study consists of 241 female respondents, which included 120 African American women (combined general population sample and oversample) and 121 White women (general population sample). Oversampling of Black residents was used, given the focus on African American women as a leading subpopulation in adverse health risks and maternal health disparities in the state and nationwide [2,3,4,5,6]. Thus, this sample is suitable for the present study to test the intersectionality hypotheses.

Our study utilized the random digit dialing method for Mississippi cellular (cell) phone numbers to establish a representative sample of adults (18 and older). Sample adult cell phone numbers were drawn from Mississippi Gulf Coast counties, specifically Hancock, Harrison, and Jackson. The random sample was supplemented by an oversample of African American adult males and females (18–50). A total of 80,000 relevant population cell phone numbers were selected from a universe of 1,083,000 numbers. A total maximum of eight dialing attempts were allotted per cell phone number before retirement. Our study utilizes data from this random sample, selecting the participant data from women who self-identified, respectively, as African American and White to form our analytical sample. Participants who were younger than 18 years of age and/or not a resident of the three coastal counties within our study parameters were informed of their ineligibility to complete the survey and discontinued from participation. Subsequently, participants were informed that they could skip responding to any item or withdraw from the survey at any time. All participants were informed of the voluntary nature of the survey and publication potential of their de-identified responses. All qualifying 419 adults completed the survey with a cooperation rate of 38.7% and 7.9%, respectively, for the general adult and oversampled African American adult populations.

In this study, the dependent variable is breastfeeding network prevalence measured by a single survey item. Respondents were asked, “How common is breastfeeding infants among your female friends and relatives?” Response categories were recoded into (0) none of them have breastfed, (1) very few of them have breastfed, (2) about one-quarter of them have breastfed, (3) about half of them have breastfed, (4) a majority of them have breastfed, and (5) all of them have breastfed. All other responses were recoded as missing values.

The key independent variable is self-reported racial identity. All responses were recoded into one variable, with African Americans = 1 and Whites = 0 (serving as the reference). In addition, household income (ranging from 1 = less than USD 10,000 to 11 = USD 200,000 or more) was utilized to measure respondents’ social class standing that may intersect with racial identity to predict breastfeeding network prevalence. Moreover, perceived community-based peer support for breastfeeding, designed to capture the sociocultural context of breastfeeding, was measured by three survey items. Respondents were asked, “How comfortable do you believe women in your community are in the following situations?” (1) nursing a baby in the presence of close women friends, (2) nursing a baby in the presence of men and women who are close friends, and (3) nursing a baby in the presence of men and women who are not close friends. All responses were recorded on a Likert scale with 1 = very uncomfortable, 2 = somewhat uncomfortable, 3 = neither uncomfortable nor comfortable, 4 = somewhat comfortable, and 5 = very comfortable. These three items were combined into a mean-score index, with higher scores representing greater perceived community-based peer support for breastfeeding. The Cronbach’s Alpha coefficient for this index was 0.683. Several sociodemographic characteristics were also included in the analysis as statistical controls: (1) age (ranging from 18 to 83), (2) educational attainment (ranging from 1 = never attended school or only attended kindergarten to 9 = beyond master’s degree), and (3) employment status (dummy-coded with not working as the reference).

Analytic Approach

The survey data were weighted to represent adults 18 years of age and older residing in the Gulf Coast region of Mississippi (Jackson, Hancock, and Harrison counties) in 2019. In addition, the data were weighted to adjust for oversampled African American males and females aged 18–50. The multiple imputation technique was utilized to estimate and replace all missing values in the dependent, independent, and control variables after confirming they were missing completely at random (MCAR) except for household income. Unless noted otherwise, all statistical analyses were conducted using the Statistical Package for the Social Sciences (SPSS) version 29.

5. Conclusions

Mississippi leads the United States in poor health outcomes, especially for African American women and infants. Breastfeeding is contingent on several social factors that have not been thoroughly investigated through the perspective of intersectionality. Our study results underscore the importance of various intersectional influences, namely, gender, race, household income (social class), and community support. Among women surveyed, racialized patterns of breastfeeding network prevalence were associated with several intersecting factors, including social class (reported household income) and the level of community breastfeeding support (perceived comfort of women breastfeeding in social situations). African American women who have higher incomes and perceive their community to be more breastfeeding-friendly are just as likely to have more breastfeeding in their social groups as White women in similar income brackets with less community support. The use of an African American oversample contributed to a unique approach in methodology to studying community perspectives and experiences related to breastfeeding. Introducing maternal healthcare policies that prioritize expanding knowledge, experience, and acceptance of breastfeeding would be a key step forward in closing racial/ethnic breastfeeding gaps. Future research would benefit the field by exploring replicable results from populations with primarily low breastfeeding initiation and duration and high infant and mother health risks. Given the intersectional reality of breastfeeding, efforts to broaden education and foster community support should enlist dynamic and tailored approaches across populations.