BMI and Disordered Eating in Black College Women: The Potential Mediating Role of Body Appreciation and Moderating Role of Ethnic Identity

Abstract

Multiple studies indicate Black American women have disproportionately higher rates of obesity compared with other groups in United States. Although body image is associated with obesity, this relation is understudied among Black women. The purpose of the current study was to (1) examine the relations among body appreciation, body dissatisfaction and disordered eating; (2) assess body appreciation as a mediator of the relation between body mass index (BMI) and disordered eating, and (3) explore ethnic identity as a moderator in this association. Participants were 191 Black women recruited from a mid-Atlantic university. Participants’ mean age was 19.16 years (SD = 1.95). Body appreciation partially mediated the relation between BMI and disordered eating behaviors and attitudes, such that women with higher BMIs reported lower body appreciation, which was associated with greater disordered eating symptoms. Ethnic identity was not a significant moderator of the association between BMI and body appreciation. Results support screening Black women with higher BMIs for disordered eating symptomatology and suggest it might be helpful to include body appreciation in interventions for Black women.

Overweight and obesity are significant public health concerns in the United States (U.S.). Nearly three-fourths (71.6%) of adults (≥20 years) have body mass indices (BMIs) classified as overweight or obese (Haua-Navarro, 2016). Moreover, multiple studies indicated that Black American women have disproportionally high levels of overweight and obesity, compared with other demographic groups (Centers for Disease Control & Prevention [CDC], 2020; Goode et al., 2018; Sutton et al., 2016). This is concerning, as obesity is associated with some of the most common causes of preventable death, including heart disease, stroke, Type 2 diabetes and certain cancers (Haua-Navarro, 2016; Must et al., 1999; Paeratakul et al., 2002). In addition, there are significant psychological and social complications of these conditions, including weight-based discrimination, depression, and anxiety (Luppino et al., 2015; Scott et al., 2008).

Eating disorders (EDs) are also a significant public health concern. Like obesity, EDs are associated with several serious medical complications (e.g., cardiac arrhythmia, osteoporosis, diabetes, infertility, asthma, bowel paralysis, electrolyte imbalance, and heart failure; Klump et al., 2009) and elevated mortality rates (Fichter & Quadflieg, 2016). In addition, more than half of all individuals with an ED have a comorbid psychiatric diagnosis such as depression or anxiety (Hudson et al., 2007). Although initially conceptualized as independent problems, EDs and obesity often have common etiologies and similar physiological and psychological comorbidities (Rancourt & McCullough, 2015). Though Black women are frequently diagnosed with obesity, concern remains that EDs in this group often go unrecognized and undiagnosed (Cachelin et al., 2000; Lydecker & Grilo, 2016).

A common risk factor for EDs is body dissatisfaction (Rancourt & McCullough, 2015). Body dissatisfaction is associated with increased body weight (Dulloo & Montani, 2015). Negative body image (i.e., body dissatisfaction) has been well explored in White populations; however, relatively few studies have examined body image exclusively in Black individuals (Antin & Hunt, 2013; Cox et al., 2011; Wood & Petrie, 2010). Although the number of studies including Black women has increased in recent decades, this work frequently examines Black women’s body image in comparison to women of other races and ethnicities (Breitkopf et al., 2007; Chithambo & Huey, 2013; Grabe & Hyde, 2006; Ordaz et al., 2018; Sutter et al., 2015). This approach precludes a thorough examination of how this construct functions within Black women specifically. Positive body image (i.e., body appreciation) in Black women has received even less attention in the literature, and little work has examined this construct in relation to BMI and disordered eating in this population.

Body Image

Body image is a complex construct defined broadly as individuals’ experiences, beliefs, and attitudes toward their body (Cash, 2004). Body image has a significant influence on weight-related behaviors and perceptions (Fitzgibbon et al., 2000; Grabe & Hyde, 2006; Kronenfeld et al., 2010; Roberts et al., 2006). The most frequently examined aspect of body image is body dissatisfaction, the negative affect and cognitions generated when there is incongruence between an individual’s actual body and her “ideal” body (Cash, 2004). The ideal body is typically a construction perpetrated by the media, family, and significant others as beautiful and/or socially acceptable (Weiss, 2000).

Body Dissatisfaction

Most research investigating body image has addressed the thin ideal which refers to an extremely thin, waifish figure; other elements of this ideal include fair skin, blonde hair, and blue eyes. In the past two decades, new research has explored the possibility of other body ideals, such as muscular (Tiggemann & Zaccardo, 2018), and curvy ideals (Capodilupo, 2015; Overstreet et al., 2010). Although these ideals differ from thinness, they are equally unrealistic and unobtainable for most women (Capodilupo, 2015) Thus, for many women, attempts to achieve these ideals elicit body dissatisfaction, which is linked to a myriad of negative health outcomes such as overweight and obesity, EDs, anxiety and depression (Bearman et al., 2006; Haines & Neumark-Sztainer, 2006; Stice & Shaw, 2002; Thompson et al., 1999).

The extant body image literature has largely focused on White women and excluded Black women. When Black women are considered, it is frequently noted that they are likely protected from body image concerns due in part to cultural reasons (Beauboeuf-Lafontant, 2003; Duke, 2000; Gentles & Harrison, 2006). Specifically, numerous researchers have asserted that because many Black Americans view a larger body size as acceptable, or perhaps even preferable, Black women do not experience thin ideal internalization, and thus are immune to body dissatisfaction (Grabe & Hyde, 2006; Milkie, 1999; Parker et al., 1995; Schooler et al., 2004). Mechanisms explaining the lower levels of body dissatisfaction in Black women are not clearly operationalized; however, recently the construct of body appreciation has garnered attention as one potential driver of the association between BMI and body dissatisfaction and disordered eating.

Positive Body Image

Positive body image was initially conceptualized as one end of a spectrum, with body dissatisfaction anchored at the other end, but current literature has revealed that this construct, although related to negative body image, is actually its own novel and complex entity (Tylka & Wood-Barcalow, 2015a). There is evidence that body appreciation contributes uniquely to outcomes such as eating disorder symptomatology (Tylka & Wood-Barcalow, 2015b).

Positive body image has at least three central components according to Menzel and Levine (2010 p. 166): “(a) appreciating the body’s appearance and function, (b) being aware of and attentive to the body’s experiences and needs, and (c) possessing a positive cognitive style for processing body-related messages in a self-protective way.” Measuring positive body image provides a foundation to develop and understand health promotion related to body image. Positive body image is related to outcomes such as intuitive eating, self-esteem, proactive coping, optimism, positive affect, self-compassion, life satisfaction, and subjective happiness (Avalos et al., 2005; Swami et al., 2014; Swami & Ng, 2015; Tylka & Kroon Van Diest, 2013). Several aspects of what is commonly described in the literature as Black culture (e.g., acceptance of a diversity of body sizes, rejection or criticism of the Eurocentric beauty standard, and an emphasis on nonphysical aspects of beauty) are encompassed within the construct of positive body image (Tylka & Wood-Barcalow, 2015a); thus, further examination of it might improve understanding of Black women’s eating attitudes and behaviors.

Multiple Beauty Ideals

The term Black culture is rarely operationalized in the body image literature, but the perceived acceptance of larger body sizes among Black women is sometimes considered a factor protecting Black women from experiencing body dissatisfaction (Capodilupo & Kim, 2014). Additionally, Black culture is often linked with both rejection or criticism of the Eurocentric beauty standard (Webb et al., 2013a), and a broader conceptualization of beauty, which includes personal style, self-care, and spirituality (Rubin et al., 2003). Generally, is it posited that individuals who are socialized to view beauty as inclusive of such characteristics are also more like to believe a curvy figure is acceptable and even desirable and to deemphasize factors such as the thin ideal (Lovejoy, 2001; Poran, 2006). In turn, the acceptance of a larger body size and rejection of the thin ideal have been reported to be protective against the development body dissatisfaction (Poran, 2006). Thus, it has been asserted that stronger identification with Black identity could potentially explain why some Black women might actively pursue a curvy beauty ideal, and/or demonstrate high levels of body appreciation relative to non-Black women (Lovejoy, 2001; Poran, 2006). Nonetheless, in other research, Black and White women manifest no significant differences in thin ideal internalization (Forbes & Frederick, 2008; Grabe & Hyde, 2006). However, ethnic identity is rarely assessed, and thus, little empirical data addressing the credibility of these conclusions are available.

Ethnic Identity

Ethnic identity is a potential moderating variable frequently cited in discussions of body dissatisfaction in Black women (Capodilupo & Kim, 2014; Hesse-Biber et al., 2010). Although the available literature has infrequently addressed the association between ethnic identity and body dissatisfaction, when this link is considered, it is generally referenced solely in relation to Black women’s identification (or lack thereof) with Eurocentric beauty ideals. Specifically, some research has indicated that Black women who identified more strongly with White culture reported higher levels of body dissatisfaction and disordered eating, which was attributed to the idea that these women were more likely to internalize the Eurocentric thin beauty ideal (Henrickson et al., 2010; Poran, 2006; Selzer, 2013).

Although some studies have found that Black women who identified strongly with Black culture ideals and standards of beauty experienced less body dissatisfaction (Oney et al., 2011; Watson et al., 2013; Wood & Petrie, 2010), several others found no direct relation between ethnic identity and body dissatisfaction (Capodilupo & Kim, 2014; Forbes & Frederick, 2008; Fujioka et al., 2009). For example, a qualitative study with Black college women provided insight into the complexity of the broad construct of ethnic identity in the context of beauty standards. Hesse-Biber et al. (2010) conducted in-depth interviews with Black college women and identified four distinct groups (i.e., White enough, Black and proud, floater, and bridge builder), with each group reporting different associations between ethnic identity and body image concerns. The “White enough” group described women who spent most of their youth and emerging adulthood with primarily White peers, identified economically as middle and upper class, and felt accepted by White peers. They reported intense pressure to be thin and engaged in restrictive eating and obsessive exercise to achieve this ideal. In contrast, the “floaters” group identified readily with both Black and White culture and reported corresponding pressures to obtain the beauty ideals of both groups, depending on the setting. Hesse-Biber et al.’s work indicated that Black women’s interpretation of their identity seemed to be influenced heavily by early experiences of socialization. Further, the degree to which women currently maintained multiethnic friend groups seemed to be related to their report of the relevance of Eurocentric beauty ideals. In sum, it is likely certain aspects of ethnic identity influence body image and disordered eating attitudes and behaviors in Black women, but in varying and complex ways. Hesse-Biber and colleagues’ work provides possible explanations to the seemingly disparate findings that some Black women endorse a preference for a larger body size, while others seem to identify more strongly with the thin ideal.

Growing evidence indicates that some Black women may reject the Eurocentric ideal of beauty, and thus, accept and appreciate higher body weights (Overstreet et al., 2010; Pickett & Peters, 2017; Ristovski-Slijepcevic et al., 2010). However, Black women are a heterogenous group and are socialized in many different environments, each of which might have different appearance expectations (Hesse-Biber et al., 2010; Sanderson et al., 2013; Selzer, 2013). Thus, it is possible that some Black women might strive to achieve the Eurocentric/thin ideal, rather than the higher weight ideal, which would likely be associated with body dissatisfaction. The measurement of ethnic identity is one potential method to understand which body ideal might be most relevant for a given Black woman. For instance, a stronger affiliation with Black identity might be associated with a greater likelihood of acceptance of higher body weight and critique of the thin ideal, which could protect against feelings of body dissatisfaction. As experiences of ethnic identity are so varied and nuanced, it is useful to explore the role of additional potential mediating and moderating variables, to enhance understanding of body image in Black women.

Purpose of the Current Study

In the current study, we examined the association among BMI, ethnic identity, body appreciation, and disordered eating behaviors and attitudes. Prior research has focused exclusively on body dissatisfaction in Black women; the current study extends previous research by examining the relation of BMI to disordered eating attitudes and behaviors in Black women, and the potential mediating role of body appreciation. Previous research has also indicated that ethnic identity might influence body image as well as eating behaviors and attitudes; thus, we also examined the potential moderating role of ethnic identity on the relation between BMI and body image. It was hypothesized that body appreciation would mediate the relation between BMI and disordered eating behaviors and attitudes in Black women. It was also hypothesized that ethnic identity would moderate the association between BMI and body appreciation, such that the association between BMI and body appreciation would be stronger at higher levels of ethnic identification.

Method

Participants

All participants identified as Black and female (N = 191), and were a subgroup from a larger study of positive body image in racially and ethnically diverse college students. Over half were in their first year with an average BMI in the overweight range. See Table 1 for details regarding demographics. Women were provided the opportunity to self-identify their race and ethnicity. The majority indicated they were “Black or African American” (83.4%, n = 165), without further specification. Other open-ended responses included African (n = 1), Arab (n = 1), Black/Native American (n = 1), Ghanaian (n = 2), Jamaican/West Indian (n = 1), Malagasy (n = 1), Multiracial (n = 2), North African (n = 2), Native American (n = 1), and Nigerian (n = 2).

Table 1.

Participant Demographics

N=191	M (SD)
Age	19.16 (1.95)
BMI	26.25 (6.59)
	% (n)
First year	59.6% (115)
Second year	21.2% (41)
Third year	11.9% (23)
Fourth year	4.7% (9)
Fifth year+	0.5% (1)
Self-identified ethnicity
Black/African American	83.4% (165)
African	0.5% (1)
Arab	0.5% (1)
Ghanaian	1.0% (2)
Jamaican/West Indian	0.5% (1)
Malagasy	0.5% (1)
Multiracial	1.0% (2)
North African	1.0% (2)
Native American	1.0% (2)
Nigerian	1.0% (2)

Note. All participants identified their race as Black, and had the option to self-identify ethnicity in an open-ended text box.

The majority indicated they were “Black or African American” (83.4%, n=165), without further specification. Open-ended responses included African (n=1), Arab (n=1), Black/Native American (n=1), Ghanaian (n=2), Jamaican/West Indian (n=1), Malagasy (n=1), Multiracial (n=2), North African (n=2), Native American (n=1), and Nigerian (n=2).

Measures

Demographics.

Participants reported their age, gender, year in school, and current height and weight, which were used to calculate body mass index with a standard formula: weight (kg)/[height (m)]² (Nuttall, 2015). Participants were also asked to self-identify their racial and ethnic identities.

Eating Disorders Self Examination Questionnaire (EDE-Q; Fairburn & Beglin, 1994, 2008).

The EDE-Q, adapted from the Eating Disorder Examination interview (EDE; Cooper & Fairburn, 1987; Fairburn & Zafra, 1993), was used to examine the frequency of disordered eating behaviors and attitudes in the past 28 days (Penelo et al., 2012). Sample questions include, “Has your weight influenced how you think about (judge) yourself as a person?” “Has thinking about shape or weight made it very difficult to concentrate on things you are interested in (for example, working, following a conversation, or reading)?” and “Have you tried to follow definite rules regarding your eating (for example, a calorie limit) in order to influence your shape or weight (whether or not you have succeeded)?” Responses are rated on scale ranging from 0 = no days/not at all to 6 = everyday/markedly. This 28 item self-report questionnaire generates four subscales (e.g., Dietary Restraint, Shape Concerns, Weight Concerns, and Eating Concerns) and a global score which represents the average response across the four subscales. The global score is calculated by taking the sum of all responses and diving it by the total number of items to create a mean total score; subscale scores are calculated by generating the sum of each item included in a given subscale and calculating the mean based on the number of items included in the subscale. Higher scores indicate greater severity and scores ≥ 4 are considered clinically significant (Luce et al., 2008; Mond et al., 2004). However, the four-factor structure has been difficult to replicate (Allen et al., 2011; Barnes et al., 2012; Grilo et al., 2013) and some researchers propose a three-factor solution in which weight and shape concern represent a single body dissatisfaction factor (Barnes et al., 2012; Peterson et al., 2007). The single body dissatisfaction subscale demonstrated superior internal consistency and was utilized in addition to the Dietary Restraint subscale in analyses. The EDE-Q has been validated specifically with Black college women (N = 395), whose mean scores were similar to those of White women from a community sample, and demonstrated internal consistency (α = .91; Kelly et al., 2012). Cronbach’s alphas for Black college women in previous literature were .91 (global score), .88 (weight concerns) .91(shape concerns) and .82 (dietary restraint; Kelly et al., 2012). The current study, as mentioned above, utilized the three-factor solution in which shape and weight concerns represented the single body dissatisfaction factor. This instrument also has been demonstrated to have strong discriminant validity, differentiating individuals with an eating disorder from those without (Aardoom et al., 2012) as well as strong convergent validity with the Eating Disorder Examination (a semi-structured interview protocol upon which the EDE-Q is based; Mond et al., 2004). Cronbach’s alphas for the current study were .93 (global score), .95 (body dissatisfaction), and .79 (dietary restraint).

Body Appreciation Scale-2 (BAS-2; Tylka & Wood-Barcalow, 2015a).

The BAS-2 was designed to assess body appreciation, the acceptance of and favorable opinions towards one’s body, and a rejection of narrow beauty ideals promoted through mass media. The BAS-2 contains 10 items rated on a 5-point scale ranging from 1 = Never to 5 = Always. Sample items include, “I respect my body,” “My behavior reveals my positive attitude toward my body; for example, I walk holding my head high and smiling.” “I am attentive to my body’s needs,” and “I feel like I am beautiful even if I am different from media images of attractive people (e.g., models, actresses/actors).” Items are averaged to derive the overall score, with higher values reflecting more body appreciation. The BAS-2 yields internally consistent scores in college women (α = .94–.97; Tylka et al., 2015; Tylka & Wood-Barcalow, 2015a) and is associated with unique variance in intuitive eating and ED symptoms (Tylka & Wood-Barcalow, 2015a). Cronbach’s alpha in this study was .95. The BAS-2 was validated in samples of 820 women and 767 men, and authors identified 11% of the sample as African American (Tylka & Wood-Barcalow, 2015a). Exploratory factor analysis with the ten items of the BAS-2 indicated that those items accounted for 64.17% of the variance (Tylka & Wood-Barcalow, 2015a). Convergent validity was also strong, as the BAS-2 was positively correlated with six other measures which examine factors such as appearance evaluation, self-esteem, intuitive eating, and proactive coping (Tylka & Wood-Barcalow, 2015a). Likewise, the instrument was negatively correlated with other instruments examining factors such as body dissatisfaction and body surveillance (Tylka & Wood-Barcalow, 2015a).

Multigroup Ethnic Identity Measure-Revised (MEIM-R; Phinney & Ong, 2007).

The MEIM-R is a 6-item self-report measure of ethnic attitudes and behaviors. Responses are rated on a 5-point rating scale ranging from 1 = Strongly Disagree to 5 = Strongly Agree. The instrument is scored by taking the mean of all responses with higher scores indicating stronger ethnic identity. Sample items include “I have a strong sense of belonging to my own ethnic group,” and “I have often done things that will help me understand my ethnic background better.” The MEIM-R yields internally consistent scores in Black (α = 0.88) undergraduates and demonstrates convergent validity with other measures of ethnic identity (Yoon, 2011). In a recent metanalytic study of the MEIM-R (N = 10,611 and 18% were identified as Black or African American) reliability coefficients were consistent across ethnic groups (Herrington et al., 2016). Cronbach’s alpha in the current study was .85. Additionally, literature examining the two-factor structure of the MEIM-R has been demonstrated a good fit with 53% of the variance shared among the two factors and the remaining unshared variance indicating independence between the two factors (Chakawa et al., 2015).

Procedure

Participants were undergraduate women ages ≥ 18 at a large, public mid-Atlantic university who were recruited by convenience sampling from the Psychology department-sponsored participant pool after ethics approval at the host institution. The current investigation was advertised as a study of attitudes and behaviors regarding body image and eating, and participants received course credit for participation. Electronic consent was obtained and surveys, lasting approximately 30 minutes, were conducted via REDCap (Research and Electronic Data Capture; Harris et al., 2009), a secure, web-based application for building and managing online databases quickly and securely. REDCap technology allows researchers to communicate securely with participants via a messaging system that was used to send study communications such as emails and survey links directly to participants. Included participants were those who answered all three validation questions correctly. Validation questions asked participants to provide specific response options, such as “choose disagree.” All participants completed the surveys in the same order. Course credit was provided as compensation for completion.

Data Analysis

A sensitivity power analysis was conducted with G*Power software (Erdfelder et al., 2009); with 80% power at α = .05, the sample of 191 participants was sufficient to detect f² ≥ .05, with a critical F of 3.04. All analyses were conducted in SPSS (IBM Corp, 2017). Data were cleaned and assumptions of planned analyses were checked, including normality, linearity, homoscedasticity, and multicollinearity.

Indirect Effect Analysis

First, we constructed a simple mediation model examining the indirect effect of body appreciation on the association between current BMI and eating attitudes and behaviors via the PROCESS macro (model 4; Hayes, 2013). BMI was entered as the predictor variable, body appreciation as the mediator, and the EDE-Q global score as the outcome variable. Two models are constructed: (1) the effect of the predictor on the mediator, controlling for age, and (2) the effect of the predictor on the outcome variable, controlling for age and the mediator (i.e., the total effect), and the effect of the mediator on the outcome variable (controlling for the predictor and age). Direct, indirect, and total effects were estimated using 5000 bootstrapped samples. If the 95% confidence interval (CI) of the indirect effect does not contain zero, we would conclude that the association between BMI and eating attitudes and behaviors is partially explained by one’s level of body appreciation.

Next, we examined a moderated mediation analysis to assess whether the indirect effect of body appreciation on the association between BMI and eating attitudes and behaviors depended on one’s level of ethnic identity (i.e., the moderator). Moderated mediation, also known as a conditional indirect effect, occurs when the effect of the predictor on the outcome variable via the mediating variable differs based on the level of the moderating variable (Hayes, 2015). For instance, if the association between greater BMI (predictor) and higher ED psychopathology (outcome variable) was partially explained by lower levels of body appreciation (mediator), this would reflect a mediational effect. If the negative association between BMI and body appreciation were weaker for those with greater ethnic identity, this would suggest a condition indirect effect, because the mediating effect changes based on the level of the moderator. We again used the PROCESS macro, with current BMI entered as the predictor, body appreciation as the mediator, and the EDE-Q global score as the outcome variable, and two models were constructed as above. We examined whether ethnic identity (i.e., the MEIM average score) moderated the association between current BMI and body appreciation (PROCESS model 7). Specifically, PROCESS was used to calculate the effect of current BMI on body appreciation at the 16^th, 50^th, and 84^th percentiles of ethnic identity. A significant interaction between current BMI and ethnic identity on body appreciation would indicate the path is moderated. Additionally, the index of moderated mediation demonstrates that the indirect effect is conditional on the moderator when the 95% CI does not contain zero (Hayes, 2015).

Results

Data Management

Data were normal (both skewness and kurtosis between −1 and 1) and outliers (z > 2.0) were minimal and not extreme; thus, all cases were retained (Meyers et al., 2016). Inspection of scatter plots revealed linear relations between study variables. Plotted predicted values and residuals supported that data were homoscedastic, and variance inflation factors were below 2.0, indicating data were not multicollinear. Thus, assumptions of planned analyses were met. Missing data were minimal (0–4 responses per item, comprising <2% of data) and missing completely at random, χ²(8) = 7.50, p = .48, rendering biased parameters unlikely (Meyers et al., 2016). Thus, we used listwise deletion, the default behavior of the PROCESS macro (Hayes, 2013). We examined whether age or year in school were associated with study variables; age was significantly correlated with the EDE-Q global score (r = .18, p = .01) and was included as a covariate. Year in school displayed no significant correlations with study variables.

Descriptive Statistics and Bivariate Correlations

Descriptive statistics and correlations are reported in Table 2. We examined bivariate correlations, which indicated the strength and direction of the relation between two variables. BMI was negatively correlated with body appreciation (r = −.38) and positively correlated with disordered eating attitudes and behaviors (r = .56), dietary restraint (r = .42), and body dissatisfaction (r = .53). Body appreciation was negatively correlated with disordered eating attitudes and behaviors (r = −.61), body dissatisfaction (r = −.69), and dietary restraint (r = −.32). The MEIM-R, which assessed ethnic identity, was not significantly correlated with any study variables.

Table 2.

Summary of Intercorrelations, Means, and Standard Deviations for Study Variables

Variable	M	SD	1	2	3	4	5	6
1. Body Mass Index	26.9	6.59	-
2. Body Appreciation	3.83	0.83	−.38**	-
3. EDEQ, Global	2.24	1.18	.56**	−.61**	-
4. Body Dissatisfaction	2.88	1.61	.53**	−.69**	.94**	-
5. Dietary Restraint	2.16	1.15	.42**	−.32**	.82**	.63**	-
6. MEIM	3.24	0.49	.03	.12	.02	.03	.01	-

Note. MEIM=Multi Ethnic Identity Measure. EDEQ = Eating Disorder Examination-Questionnaire.

^**p < .01.

Indirect Effect Analysis

An indirect effect analysis examined the indirect effect of BMI on eating behaviors through body appreciation, controlling for age. Results are presented in Figure 1. BMI was negatively associated with body appreciation and positively associated with disordered eating attitudes and behaviors (both p < .001). Body appreciation was negatively associated with disordered eating attitudes and behaviors (p < .001). Because the 95% confidence interval did not include zero (0.02–0.05), we concluded that the positive relation between BMI and disordered eating attitudes and behaviors was partially explained by lower body appreciation.

Figure 1.

Moderated Mediation Model of Associations Between Eating and Weight Related Constructs.

Note. This is a moderated mediation model for the indirect effect of body appreciation on the association between BMI and eating disorder (ED) symptoms, and the moderating effect of ethnic identity on the association between BMI and body appreciation. Age was included as a covariate. Unstandardized coefficients are presented. Gray, dashed lines indicate nonsignificant paths. *p < .05. **p < .001.

Moderated Mediation Analysis

The PROCESS macro involved two models to assess moderated mediation. The first model assessed the associations among the predictor, mediator, and moderator. The second model assessed the effects of the predictor, mediator, and moderator on the outcome. In the first model, the overall model was significant: F (3, 167) = 12.29, p < .001, r² = .18. However, none of the individual effects were significant. Specifically, the associations between the covariate (age) and body appreciation (b = −.03, SE = .03, p = .25), BMI (predictor) and body appreciation (mediator; b = −.03, SE = .06, p = .62), and ethnic identity (moderator) and body appreciation (b = .42, SE = .50, p = .40) were not significant. Similarly, the interaction effect of BMI and ethnic identity on body appreciation (b = −.01, SE = .02, p = .70) was not significant.

The second model was also significant overall, F (2, 168) = 72.88, p < .001, r² = .46. There was a positive association between the covariate of age and EDE-Q global scores (b = .06, SE = .03, p < .05). Both current BMI (b = .07, SE = .01, p < .001) and body appreciation (b = −.68, SE = .09, p < .001) were significantly related to global EDE-Q scores, reflecting that those with higher current BMI and lower body appreciation reported greater disordered eating attitudes and behaviors. However, the index of moderated mediation (b = .005, SE = .01, 95% CI [−.02, .04]) supported the lack of a moderated mediation (see Figure 1), indicating that regardless of ethnic identity level, the association between higher BMI and greater global EDE-Q scores was partially explained by lower body appreciation in our sample.

Discussion

The purpose of the current study was to examine relations among BMI, body appreciation, body dissatisfaction, and disordered eating behaviors and attitudes in Black women. Descriptive results indicated higher BMI is associated with greater body dissatisfaction, dietary restraint, and global eating pathology. Although these results are cross-sectional, they offer important empirical support to the growing body of literature suggesting that, contrary to some earlier hypotheses, (Grabe & Hyde, 2006; Molloy & Herzberger, 1998; Parker et al., 1995), Black women do experience body dissatisfaction and disordered eating. Specifically, current data provided evidence that Black women’s body dissatisfaction and disordered eating behaviors are associated with their body weight. Thus, these findings add to the small body of work highlighting that Black women may not be immune to weight and body image concerns (Capodilupo, 2015; Forbes & Frederick, 2008; Wood & Petrie, 2010). Furthermore, these findings indicated that body weight maybe a relevant factor to consider when assessing Black women’s body dissatisfaction and engagement in disordered eating behaviors.

In addition, body appreciation partially mediated the relation between BMI and body dissatisfaction; individuals with higher BMIs had less body appreciation, and higher levels of disordered eating attitudes and behaviors. Most previous research suggests that Black women rejected the thin ideal, because it was inconsistent with a cultural preference for larger bodies (Awad et al., 2015; Hesse-Biber et al., 2010; Pickett & Peters, 2017; Webb et al., 2013a). These results led to the assumption that Black women embrace higher body weights; it has even been asserted that Black women’s acceptance of a higher body weight underlies the “obesity crisis” among Black women (Pickett & Peters, 2017). In contrast to this assumption, current findings demonstrated that Black women with higher body weights had lower body appreciation, and higher body dissatisfaction and disordered eating behaviors. These findings support the evidence that Black women do not universally accept higher body weights. Indeed, a growing body of work has demonstrated that many Black women desired weight loss, expressed unhappiness with their current weight, and engaged in disordered eating behaviors to control their weight (Capodilupo, 2015; Selzer, 2013).

Both the current results and prior research with Black women might be explained, in part, by the fact that body weight and shape are relevant to a variety of beauty ideals. For example, the essential features of the curvy ideal include voluptuous bust and hips, accompanied by a very slim waist (Antin & Hunt, 2013; Overstreet et al., 2010). Overstreet et al. (2010) reported that Black women who endorsed the curvy ideal manifested higher levels of body dissatisfaction, which was related to weight dissatisfaction, shame, and body monitoring. Thus, although the curvy ideal differs from the thin ideal, it nonetheless renders those who endorse it vulnerable to body dissatisfaction.

At the same time, there is emerging evidence that some Black women aspire to the thin ideal, similar to many of their White counterparts (Forbes et al., 2005; Grabe & Hyde, 2006; Roberts et al., 2006; Shaw et al., 2004). Some literature suggests that as diet culture has continued to permeate into the mainstream, Black women have become increasingly vulnerable to internalization of the thin ideal (Webb et al., 2013b, 2014). Furthermore, Black women are portrayed more frequently in mainstream media, but often in thin and/or athletic bodies (Quinlan et al., 2012). Likewise, Black women have reported that even when they do not have a personal preference for the thin ideal, they recognize that there is a benefit to having a figure more closely aligned with the Eurocentric beauty standard (Awad et al., 2015; Dickens & Chavez, 2018). Thus, in sum, Black women are faced with the “double disadvantage” of navigating both Eurocentric beauty expectations and within-group ideals, which are both unobtainable, and often times, contradictory (Blair et al., 2005; Kilbourne, 1994; Plybon et al., 2009).

The current study also found that Black women with higher BMIs have lower levels of body appreciation, which in turn, was related to greater body dissatisfaction and disordered eating behaviors. Interestingly, although the literature has long focused on the idea that Black women reported less body dissatisfaction than their White counterparts, there has been a relative paucity of research on positive body image in Black women. An essential component of body appreciation is the cognitive flexibility to reject unrealistic or negative beauty ideals, consistent with the cognitive behavioral model of body image (Tylka & Wood-Barcalow, 2015b; Webb et al., 2014). Individuals with lower body appreciation are likely less flexible regarding these beauty ideals. This, in turn, could render them more vulnerable to body dissatisfaction resulting from the inability to achieve these ideals, which in turn increases susceptibility to disordered eating attitudes and behaviors.

The current study also examined the role of ethnic identity as a potential moderator of the association between BMI and body appreciation. Contrary to our hypothesis, we did not find evidence of ethnic identity as a moderator in our analyses. Previous studies have yielded conflicting findings about the influence of ethnic identity on body image in Black women. There is evidence which suggests that other constructs, such as internalization of the thin ideal, might be more relevant in explaining body image concerns in this group rather than ethnic identity (Forbes et al., 2005; Grabe & Hyde, 2006; Roberts et al., 2006; Shaw et al., 2004). Similarly, there may be other mechanisms, either at the developmental level, such as early socialization (Williams et al., 2013), or contextual factors, such as professional appearance expectations (Rosette et al., 2018), which might better explain Black women’s potential level of identification with or rejection of the Eurocentric beauty ideal. In addition, there was a restricted range in MEIM-R scores in the current sample, which might have limited our ability to observe an interaction.

Strengths

This is the first known study to explore the role of body appreciation in the relation between BMI and disordered eating behaviors and attitudes in Black women. Furthermore, it expands previous literature demonstrating Black women are not immune to either body dissatisfaction or disordered eating behaviors, contrary to earlier work in the field. In addition, this study is one of few quantitative investigations that exclusively examined within-group differences in Black women, rather than comparing them to White women or grouping them with other women of color. Lastly, the current study included negative and positive body image as distinct constructs, rather than as opposite ends of a spectrum.

Limitations

This study was conducted with a relatively small sample of college women enrolled in a predominately White (55%) institution. Because Black college students are a specific subset of a heterogenous group, results from the current study should not be generalized to all Black women. Although college aged Black women face novel risk factors with regard to eating and weight concerns (Gokee-LaRose et al., 2009), future work should be conducted with larger, more age and educationally diverse community samples. Importantly, issues related to body image as well as eating and weight-related behaviors change across the lifespan; thus, examination of similar constructs with women at various developmental stages is critical for improved generalizability. Further, we did not collect socioeconomic data, which would enhance interpretations of generalizability in the future. In addition, although the most widely used measure of body appreciation was included, other aspects of positive body image were not studied; a mixed method approach would allow for a more in-depth examination of this construct. Moreover, we examined ethnic identity with the MEIM-R, a brief instrument that may not fully capture the behaviors and attitudes most central to understanding Black women’s body image. Likewise, our sample size may not have been large enough to adequately represent the heterogeneity in ethnic identity that is representative of the broader population of Black women. Thus, we do not know the degree of potential heterogeneity within this sample of Black women. Finally, the surveys were presented in a consistent order; thus, results might have been influenced by order effects.

Future Directions

Future research should use measures that could capture more nuanced aspects of identity such as the Black Racial Identity Attitude Scale (Helms & Parham, 1996) and include larger samples of Black women of different ages and from different educational statuses. Furthermore, the current study examined only body weight and not body shape. Although there is evidence supporting the validity of self-reported height and weight in undergraduate populations (Quick et al., 2015), future studies should include a measure of body shape to provide a more comprehensive understanding of body image and ideals in Black women. Similarly, measures should be included that can more readily assess the full range of body modification practices, including behaviors which might be done to achieve other body ideals outside of thinness, such as the curvy ideal.

Clinical Implications

Black women often receive the message that they should diet based on their BMI status (Thomas et al., 2008). Although obesity is a serious health issue, which is associated with numerous comorbidities, (Haua-Navarro, 2016; Must et al., 1999; Paeratakul et al., 2002), there are concerns about the appropriateness of the BMI metric to categorize Black women as being “over-weight” or having “obesity;” notably, that it differentially captures adiposity in Black individuals compared to those from other demographic groups (Heymsfield et al., 2016; Rahman & Berenson, 2010). In particular, younger Black women have greater fat free mass than their White peers with equivalent BMIs (Heymsfield et al., 2016). Moreover, messages promoting dieting can inadvertently contribute to weight stigma, which is itself associated with weight gain (Tomiyama et al., 2018).

Rather than continuing to design interventions for Black women exclusively for weight loss (BMI reduction), clinicians might consider developing holistic strengths-based interventions that include factors, such as body appreciation and physical activity, which promote attunement and body appreciation. Indeed, weight loss might be important for many Black women in situations in which it contributes to comorbid health conditions, or impairs other aspects of physical or psychological functioning; however, it is important to consider the etiology of weight concerns when making treatment recommendations. For example, although caloric restriction can be an effective and important aspect of behavioral weight loss, it might not be the most appropriate treatment recommendation for an individual experiencing intrusive and distressing thoughts related to eating and weight or engaging in restrictive eating. Thus, when Black women present in healthcare settings with BMIs in the overweight or obese category, clinicians should assess for disordered eating behaviors and cognitions prior to making any weight loss recommendations.

Conclusion

The current study found Black women with higher BMIs reported lower body appreciation, and higher body dissatisfaction, eating restraint, and global eating disorder symptoms. These findings support the extant literature demonstrating Black women evidence body dissatisfaction and disordered eating at rates comparable to those of White women. This information is important, as early literature suggested Black women were protected from eating pathology and body dissatisfaction due in part to cultural ideals. These data also highlight the need for further research on this understudied topic, given the contradictory findings regarding Black women’s body image in the extant literature. In addition, future research should continue to examine body image related constructs using within-group samples, as a major limitation of the current literature is the focus on comparing Black women to White women, who are considered the normative group. In conclusion, clinicians should consider the role of disordered eating in Black women’s body weight and body image concerns. Future interventions with Black women should build on these findings and emphasize the importance of body appreciation.