Racial Discrimination and Low Household Education Predict Higher Body Mass Index in African American Youth

Abstract

Objective:

The purpose of this study was to examine the relationships between environmental factors, including household education, community violence exposure, racial discrimination, and cultural identity, and BMI in African American adolescents.

Methods:

A community-based sample of 198 African American youth (120 girls, 78 boys; ages 11–19 years) from Washtenaw County, Michigan, were included in this analysis. Violence exposure was assessed by using the Survey of Children's Exposure to Community Violence; racial discrimination by using the Adolescent Discrimination Distress Index; cultural identity by using the Acculturation, Habits, and Interests Multicultural Scale for Adolescents; and household education by using a seven-category variable. Measured height and body weight were used to calculate BMI.

Results:

Racial discrimination was positively associated with BMI, whereas household education was inversely associated with BMI in African American adolescents (discrimination: β = 0.11 ± 0.04, p = 0.01; education: β = −1.13 ± 0.47, p = 0.02). These relationships were significant when accounting for the confounding effects of stress, activity, diet, and pubertal development. Significant gender interactions were observed with racial discrimination and low household education associated with BMI in girls only (discrimination: β = 0.16 ± 0.05, p = 0.003; education: β = −1.12 ± 0.55, p = 0.045). There were no significant relationships between culture, community violence exposure, and BMI (all p's > 0.05).

Conclusion:

Environmental factors, including racial discrimination and low household education, predicted higher BMI in African American adolescents, particularly among girls. Longitudinal studies are needed to better understand the mechanisms by which these environmental factors increase obesity risk in African American youth.

Introduction

Pediatric obesity continues to exact a considerable public health toll, particularly among ethnic minority youth.¹ The most recent National Health and Nutrition Examination Survey (NHANES) estimates from 2012 reported that 40% of African Americans, ages 12–19, were overweight or obese compared with 31% of non-Latino whites.¹ Although several behavioral mechanisms have been identified to explain the greater obesity prevalence among African American youth,² the independent role of the social environment remains less clear.

Household factors, including low levels of household education, have been associated with increased obesity risk in children and adolescents.³ Data from the National Center for Health Statistics suggest that 12% of boys living in households where the head of the household had at least a college degree (i.e., high education) were obese compared with 21% of those living in households where the head of the household had less than a high school degree (i.e., low education).⁴ Among girls, 8% of children and adolescents in high-education households were obese compared with 20% in low-education households.⁴ More specifically, among African American girls, the prevalence of obesity was 17% in high-education households compared with 26% in low-education households.⁴ However, the protective effects of household education were not observed for African American boys.⁴ Others have found no relationships⁵ or positive associations^6,7 with parent education and obesity in ethnic minority youth, warranting further research.

Neighborhood factors, including community violence exposure, can also impact obesity risk via decreased participation in physical activity.⁸ For youth living in neighborhoods where there is a greater prevalence of community violence, fear of walking in one's neighborhood or low level of trust in one's neighbors may lead to increased psychological stress⁹ and/or decreased physical activity participation outdoors.¹⁰ Indeed, focus groups conducted with African American mothers suggested that the threat of violence strongly influenced the amount of daily outdoor physical activity in which their daughters participated.¹¹ Hence, perceptions of crime and lack of safety within a community can be a barrier to meeting physical activity recommendations, thereby potentially increasing obesity risk in African American youth.¹⁰

Psychosocial factors, including increased exposure to racial discrimination, can indirectly increase obesity risk via stress-induced changes in food consumption. Brodish et al. demonstrated that racial discrimination experienced during adolescence was associated with decreased healthy eating at age 30 in a sample of middle-class African American and non-Latino white men and women.¹² In response to an acute laboratory stressor, Bermudez-Millan et al. noted in middle-aged African American and non-Latino white women that exposure to higher levels of racial discrimination predicted more eating and alcohol consumption after the stress condition when they were exposed to a stressor compared with when they were not.¹³ These findings suggest that exposure to racial discrimination is associated with stress eating and subsequent obesity in adulthood. These findings, however, are not universal as previous research examining the relationship between racial discrimination and BMI has yielded mixed results in adults,^14–16 and fewer researchers have examined these relationships during adolescence.

Finally, sociocultural factors can play an important role in shaping obesity risk via increased psychological stress and subsequent changes in food consumption in African American youth. With the rapid increase in the ethnic and cultural diversity of the United States, African American cultures are evolving and quickly becoming a part of mainstream American culture.¹⁷ These distinct cultures have aspects that are uniquely shaped by historical, social, political, and economic forces present within the United States.¹⁸ Consequently, African American children and adolescents who come of age in the United States interact with people from different cultural backgrounds, which can lead to an interchange of cultural attitudes, beliefs, and behaviors.¹⁷ Previous research suggests that adolescents who undergo or pursue cultural integration—combining aspects of their family's culture with aspects of mainstream American culture—experience less stress and achieve better health outcomes.¹⁹ However, adolescents who have difficulty integrating into society may suffer from greater levels of stress, which, in turn, may result in changes in food consumption to cope with that stress.²⁰ Few researchers have examined the relationships between cultural integration and obesity risk in African American youth.

Taken together, social environmental factors, such as low household education, community violence, exposure to racial discrimination, and cultural identity, can contribute to increased obesity risk, either directly or indirectly through increased psychological stress and/or changes in obesity-related behaviors, including decreased physical activity and increased dietary intake.^21,22 Despite these well-characterized relationships, few researchers have attempted to disentangle these associations to better understand the increased obesity risk reported in African American youth. Because adolescence is a particularly sensitive period of life when the social environment can have lasting effects on obesity outcomes into adulthood,²³ we sought to examine the relationships between household education, community violence, racial discrimination, acculturation, and BMI in African American youth. It was hypothesized that greater exposure to violence and racial discrimination as well as lower household education and cultural integration would be associated with increased BMI in African American adolescents.

Materials and Methods

Sample Population, Recruitment, and Procedure

Children and adolescents were recruited from neighboring communities in southeast Michigan to participate in the Health & Culture Project, a nutrition and physical activity education program that examined the social, cultural, and psychological predictors of obesity. A total of 273 participants, ages 11–18 years, were recruited from middle schools (n = 85), high schools (n = 31), afterschool programs (n = 46), community centers (n = 30), and the general community in Washtenaw County, Michigan (n = 81), via flyers and word of mouth. Data collection occurred at the Childhood Disparities Research Laboratory or off-site at a school, after-school program, or community center. A subset of 198 African American adolescents (120 girls, 78 boys) who completed questionnaires assessing community violence exposure, racial discrimination, cultural integration, psychological stress, food intake, and had at least 4 days of valid accelerometer data were included in this analysis. Participants were excluded from the sample for the following reasons: 21 participants completed less than 50% of the surveys, 19 participants did not self-identify as African American or Black, 1 participant did not have recorded BMI data, and 30 participants were outliers for caloric intake (i.e., <500 and >3500 kcal/day²⁴). Because under-reporting of total calories is a common occurrence among obese adolescents and adults,²⁵ we made additional efforts to minimize this problem. The Harris-Benedict equation was used to calculate estimated total daily energy requirements. These values were then compared with children's recorded caloric intake. Using recommended criteria,^24,26 we excluded children who reported consuming less than 25% of estimated energy requirements (under-reporters) or reported consuming more than 175% of estimated energy requirements (over-reporters) from the analysis (n = 5).

In total, 75 participants were excluded from this analysis. When participants who were included (n = 198) were compared with those who were excluded (n = 75), both groups were similar in BMI, height, and gender distribution (data not shown, p's > 0.05). Included participants, however, were older (14.7 vs. 13.8 years old; p < 0.01) and heavier (66.0 vs. 53.2 kg; p < 0.01) compared with excluded participants. Written consent from the child and their caregiver (if the child was younger than 18) was obtained before participating. This study was reviewed and approved by the University of Michigan Institutional Review Board.

Household education

Household education was assessed by using a seven-category variable: (1) less than 8th grade, (2) finished 8th grade, (3) some high school, (4) high school graduate/general education development, (5) some college or vocational school (at least 1 year), (6) college graduate, or (7) graduate or professional training. Caregivers of participants completed the questionnaire to determine household education. For families with multiple caretakers (n = 135), the highest education score was used.

Community violence

Community violence exposure was measured by using the Survey of Children's Exposure to Community Violence.²⁷ This four-category survey asked whether the participant had witnessed someone being hit/punched, witnessed someone get attacked with a knife, heard a gunshot, or saw someone get shot. To follow up, the participants were asked whether they knew the victim of the violent attack and whether they had witnessed each type of violence more than once. The questions were scored on a three-point frequency scale (0 = never happened; 1 = happened but only once; and 2 = happened more than once). This questionnaire has been validated in African American children 8 years of age and older.²⁷ This measure demonstrated high reliability in the present analysis (Chronbach's α = 0.78).

Racial discrimination

Racial discrimination was measured by using the Adolescent Discrimination Distress Index.²⁸ This 15-item survey was used to assess whether participants had experienced a certain type of discrimination (educational, institutional, and peer). An example of educational discrimination was being wrongly disciplined or given after-school detention. An example of institutional discrimination was being hassled by a store clerk or store guard. An example of peer discrimination was being called racially insulting names. Each type of discrimination was summed to calculate the total number of discriminatory events experienced. This measure demonstrated high reliability in the present analysis (Chronbach's α = 0.86) and has been validated in African American adolescents.²⁸

Acculturation

Acculturation was measured by using the Acculturation, Habits, and Interests Multicultural Scale for Adolescents (AHIMSA).²⁹ Participants were asked to answer eight questions regarding their culture, such as “I am most comfortable being with people from…” The response categories were, “The United States” (indicating assimilation), “The country my family is from” (indicating separation), “Both” (indicating integration), and “Neither” (indicating marginalization). The most frequent response for each participant was then assigned a score, with integration = 0 and all others cultural orientations = 1. This questionnaire has been validated in a multi-ethnic sample of children and adolescents²⁹ and demonstrated high reliability in the present analysis (Chronbach's α = 0.87).

Body mass index

BMI and BMI percentiles were calculated by using guidelines from the Centers for Disease Control and Prevention (CDC 2000). Height (cm) and body weight (kg) were measured to the nearest 0.1 cm and nearest 0.1 kg, respectively, by using an electronic scale (Doran Scales, Inc., Batavia, IL) and ShorrBoard^® (Weigh and Measure, LLC, Olney, MD) by using standardized procedures.³⁰

Given the moderating effects of pubertal development, physical activity, dietary intake, and psychological stress on the relationships between environmental factors and BMI,^3,10,23,31 these variables were also included in the regression models. Pubertal development was measured by using the Pubertal Development Scale.³² This measure demonstrated high reliability in the present analysis (Chronbach's α = 0.80) and has been validated among adolescents.³³ Moderate-to-vigorous physical activity (MVPA) was measured by using GT3X ActiLife accelerometers. Puyau et al. cut-points, which are commonly used in adolescent populations,³⁴ were used to derive activity intensities (i.e., MVPA).³⁵ Total daily caloric intake was collected by using the 2012 Adolescent Food Frequency Questionnaire (FFQ) developed by Harvard University.³⁶ This questionnaire was designed, validated, and calibrated for use in adolescents.³⁶ Psychological stress was assessed by using the Perceived Stress Scale, which has been validated and widely used in both adolescent and adult populations.³⁷ This measure demonstrated high reliability in the present analysis (Chronbach's α = 0.80).

Statistical Analyses

Analyses were conducted by using Stata Special Edition 14.0. Multiple imputation was used to account for missing data. The majority of study variables had <10% missing data. The highest fractions of missing data were daily caloric intake (32.0%), physical activity (36.0%), and household education (34.0%). When participants who had household education (n = 130) were compared with those without household education (n = 68), there were no significant differences in gender distribution (63.1% vs. 55.9% girls; p = 0.33), BMI (25.2 vs. 24.0 kg/m²; p = 0.25), height (163.0 vs. 163.3 cm; p = 0.80), and weight (67.0 vs. 64.2 kg; p = 0.31). Those with household education information, however, tended to be younger (14.9 vs. 14.3 years, p = 0.05). Once the imputation was complete, regression was used to examine the relationship between independent variables (household education, community violence exposure, racial discrimination, and cultural integration) and BMI in African American adolescents, with pubertal status, dietary intake, physical activity, and psychological stress included as covariates. Because previous research has identified gender differences in the relationships between social environmental factors and BMI,^38,39 a second regression model was run with gender included as an interaction term. The significance level for all analysis was α < 0.05.

Results

Participant characteristics are displayed in Table 1. The average age of participants was 15 years, with 71% classified as late- or post-pubertal and 38% classified as overweight or obese. Approximately 55% of the caregivers of child participants reported having at least a college degree. Approximately 45% of participants reported that the majority of their cultural beliefs, behaviors, and attitudes came from both the United States and their families' culture (i.e., integration). In addition, 83% of the participants surveyed had experienced some form of racial discrimination in the past 6 months and 82% had been exposed to some form of community violence. Consistent with the national averages for African American children and adolescents, participants in this study engaged in 20 minutes of MVPA per day.⁴⁰ Caloric intake was somewhat lower in the present sample compared with national averages (1800 vs. 2100 kcal/day).⁴¹

Table 1.

Participant Characteristics

	n	Mean ± SE	Percentage
Sex	198	—	100
Female	120	—	61
Male	78	—	39
Pubertal development	197	—	100
Prepubertal	2	—	1
Early pubertal	11	—	6
Midpubertal	44	—	22
Late pubertal	122	—	62
Post-pubertal	18	—	9
Age (years)	198	14.7 ± 0.14	—
Weight (kg)	198	66.0 ± 1.31	—
Height (cm)	198	163.1 ± 0.67	—
BMI (kg/m²)	198	24.8 ± 0.47	—
BMI percentile	198	73.0 ± 1.87	—
Parent education	130	—	100
Less than 8th grade	1	—	1
Finished 8th grade	0	—	0
Some high school	6	—	5
High school graduate/GED	17	—	13
Some college or vocational school	34	—	26
College graduate	39	—	30
Graduate or professional training	33	—	25
Cultural integration	182	—	55
Racial discrimination score	176	13.1 ± 0.99	—
Community violence exposure score	183	0.36 ± 0.02	—
Perceived stress	192	25.4 ± 0.44	—
MVPA (min/day)	126	20.3 ± 1.31	—
Dietary intake (kcal/day)	137	1832 ± 61.85	—

Data are mean ± SE.

GED, general education development; MVPA, moderate-to-vigorous physical activity; SE, standard error.

Table 2 shows the standardized β coefficients for the associations between household education, community violence exposure, racial discrimination, and cultural integration. Household education was negatively associated with BMI (p = 0.02), whereas racial discrimination was positively associated with BMI (p = 0.01). Pubertal development was positively associated with BMI (p < 0.01). There was a trend toward a significant inverse association between physical activity and BMI (p = 0.05). There were no significant associations between community violence exposure, cultural integration, and BMI in the present analysis (p's > 0.05). There were also no significant associations between dietary intake, psychological stress, gender, and BMI (p's > 0.05).

Table 2.

Standardized Beta Coefficients for the Associations between Household Education, Community Violence Exposure, Racial Discrimination, Cultural Integration, and BMI

BMI	β	SE	p	95% CI
Parent education	−1.13	0.47	0.02^*	−2.06, −0.20
Community violence exposure	−0.48	1.74	0.78	−3.92, 2.96
Racial discrimination	0.11	0.04	0.01^*	0.02, 0.19
Acculturation	0.75	0.96	0.44	−1.14, 2.64
Caloric intake	−0.001	0.001	0.24	−0.003, 0.001
MVPA	−0.09	0.05	0.05^*	−0.18, −0.001
Perceived stress score	0.08	0.08	0.30	−0.07, 0.24
Sex	−0.43	1.28	0.74	−2.97, 2.11
Pubertal development	2.74	0.71	<0.0001^*	1.34, 4.14

Standardized beta coefficients are shown.

p < 0.05.

Table 3 shows the standardized β coefficients for the gender interactions for the relationships between BMI, household education, community violence exposure, racial discrimination, and cultural integration. Significant gender interactions were observed with racial discrimination and low household education associated with BMI in girls only (discrimination: p = 0.003; education: p = 0.045).

Table 3.

Predictors of Body Mass Index Stratified by Gender

Predictors	β	SE	p	95% CI
Parent education
Boys	−1.29	0.74	0.10	−2.73, 0.23
Girls	−1.12	0.55	0.045^*	−2.22, −0.03
Boys vs. girls	0.13	0.89	0.89	−1.64, 1.89
Community violence
Boys	−1.94	2.54	0.45	−6.95, 3.07
Girls	1.90	2.44	0.44	−2.92, 6.73
Boys vs. girls	3.85	3.51	0.28	−3.08, 10.77
Racial discrimination
Boys	0.05	0.07	0.49	−0.09, 0.18
Girls	0.16	0.05	0.003^*	0.05, 0.26
Boys vs. girls	0.11	0.08	0.60	−0.05, 0.26
Acculturation
Boys	1.60	1.47	0.28	−1.29, 4.50
Girls	0.09	1.24	0.94	−2.35, 2.53
Boys vs. girls	−1.51	1.88	0.42	−5.24, 2.21

Standardized beta coefficients and SEs are shown.

p < 0.05.

Discussion

This study sought to examine the relationships between household education, exposure to community violence and racial discrimination, cultural integration, and BMI in African American youth. Consistent with our hypotheses, residing in a low-education household and exposure to racial discrimination were associated with higher BMI in our cohort of African American adolescents, particularly among girls. These relationships remained significant even when accounting for the confounding effects of psychological stress, caloric intake, physical activity, and pubertal development. Inconsistent with our hypotheses, community violence exposure and cultural integration were not associated with BMI in this group. Taken together, these findings suggest that social and environmental factors may play an important role in shaping obesity outcomes in African American youth.

A protective effect of educational attainment on obesity risk has been well established among non-Latino white youth; however, mixed results have been observed in ethnic minority youth.^5,6 Using data from the National Health and Nutrition Examination Survey and National Health Interview Survey, Sobal and Stunkard⁵ reported that ethnic minority children, including African American youth from higher socioeconomic (income and education) households, were just as likely to be overweight and obese as compared with children residing in lower socioeconomic households. Data from the National Longitudinal Study of Adolescent Health reported a higher prevalence of obesity with increasing socioeconomic status (income and education) among African American females. Among males, this obesity disparity was the lowest at the average socioeconomic level.⁶ In contrast, Troiano and Flegal reported a pattern of decreasing overweight prevalence with increasing household education among African American girls.⁷ However, for African American boys, the lower overweight prevalence was observed with ≥13 and ≤11 years of education, but not with 12 years of household education.⁷ More recently, Ogden et al. reported a lower prevalence of obesity among African American girls residing in households headed by individuals with a college degree than in households headed by individuals with less than a high school degree; this relationship was not present in African American boys.¹ Consistent with the Ogden findings, a protective effect of educational attainment was observed in African American girls in this study.

One plausible explanation for the link between household education and obesity is the mediating role of household education in children's health beliefs and behaviors. Having more education can lead to both general and specific health-related knowledge and better problem-solving skills to make more informed health decisions.³ Household education is also closely linked with employment opportunities and family income, which, in turn, enable families to live in safer and healthier neighborhoods with supermarkets, parks, and places to exercise.³ Family income provides parents with the ability to purchase health-related goods, whereas parents' education informs health decisions, including those related to obtaining and managing medical care.³ All of these resources associated with educational attainment enable parents/caregivers to create healthy home environments and model healthy behaviors for their children. Future research should seek to better understand which specific dimensions of household education (e.g., health-related knowledge, problem-solving skills, and informed decision making) are associated with lower BMI in both African American youth.

Few researchers have examined the relationship between BMI and racial discrimination in African American adolescents. Research conducted in Asian American youth suggested that exposure to racial discrimination was a predictor of BMI in this ethnic minority group.⁴² Research conducted in black adolescent children aged 14–16 years on the Caribbean island of Barbados showed that BMI and waist circumference were greater for black girls with high levels of internalized racism and hostility compared with those with low levels of both variables; this relationship was not observed in the boys.⁴³ In this study, we also observed a significant relationship between exposure to racial discrimination and BMI among African American girls. Although it is often documented that racial discrimination is associated with increased psychological stress and increased food intake,^12,13 leading to increased obesity, the relationship between exposure to racial discrimination and BMI was significant even when accounting for these factors in the present analysis, pointing to a potential physiological mechanism. It is plausible that exposure to racial discrimination may trigger the hypothalamic-pituitary adrenal (HPA) axis response, leading to an increase in basal and stress-induced cortisol levels.⁴⁴ This activation of the HPA axis can occur even in the absence of perceptions of heightened stress. As a result, prolonged glucocorticoid exposure can lead to the development of central obesity over time. Consistent with this hypothesis, race-based discrimination was associated with greater visceral fat in African American and Caucasian women.⁴⁵ Future research should examine the interplay between racial discrimination, glucocorticoid responses, and obesity in African American youth.

Of note, the majority of race-based discriminatory events experienced by African American youth tended to stem from discrimination occurring at the institutional level, followed by educational discrimination, and lastly peer discrimination. Similarly, Fisher et al. noted that African American youth face greater instances of institutional discrimination in stores or by police compared with other ethnic minority youth.²⁸ Taken together, these findings provide insight into which type of racial discrimination may contribute to higher BMIs in African American youth and point to a potential avenue for intervention.

Studies have reported mixed results in the relationships between racial discrimination and obesity,^{16,39,42,46–48} weight gain,¹⁴ waist circumference,^48–50 and waist-to-hip ratio¹⁵ among African American adults. Data from the Black Women's Health Study, a prospective follow-up of U.S. black women aged 21–69 years, demonstrated that everyday and lifetime exposure to racial discrimination was associated with increased weight gain.¹⁴ In contrast, another study among a sample of black women aged 35 to 49 years found that high levels of perceived racism were associated with lower waist-to-hip ratio.¹⁵ Further, another study among a multi-ethnic population-based sample of U.S. adults reported a positive association between perceived chronic discrimination and waist circumference among a sample of Irish, Jewish, Polish, and Italian whites, but not among blacks, Hispanics, or other whites.¹⁶

Gender differences in the relationship between racial discrimination and obesity have also been observed, again with mixed results. Data from the 1995–2004 Midlife Development in the United States (MIDUS) cohort study found that perceived interpersonal everyday discrimination was associated with an increase in waist circumference over time, and this relationship was stronger in men than in women.¹⁶ However, data from the CARDIA study found that an increase in self-reported experiences of racial discrimination over time was significantly associated with an increase in waist circumference and BMI among black women only.³⁸ Residence in a neighborhood with high black racial isolation (i.e., residential segregation) has been shown to be associated with a higher BMI among women, but not men.³⁹ Future research is needed to better understand gender differences in the relationship between BMI and racial discrimination.

Although 82% of the participants in this study had been exposed to community violence, this environmental factor was not associated with BMI. Similarly, Gooding et al. noted that the associations between community violence and emotional abuse history with BMI were not significant after adjustment for race/ethnicity and household education.⁵¹ Gooding et al. did, however, report that participants who had witnessed domestic violence were six times as likely to be obese compared with those who had not witnessed violence in the home.⁵¹ It is likely that domestic violence (a type of violence that is of a more repetitive nature and typically takes place in the home) is a more stressful form of violence exposure compared with community violence, thereby potentially having a larger effect on adolescent health, and in this case BMI. Future research can benefit from more qualitative research to better understand how specific types of violence (domestic vs. community) may contribute to increased obesity risk in this ethnic minority population.

African Americans are a heterogeneous ethnic group who vary in the extent to which they both retain their black-American culture and also adopt aspects of mainstream white-American culture.⁵² Previous research has documented the relevance of these adaptive cultural styles to health and health-related behaviors in both African American adults and youth.^53–55 Dressler et al. reported that cultural consonance (living in accordance with culturally constructed local community norms) was a stronger independent predictor of hypertension than compared with indicators of socioeconomic status in African American adults.⁵³ Airhihenbuwa et al. reported that positive identification with African American culture and a self-perception of being successful in both the “black” and “white” ways of life were associated with reduced fat consumption and more participation in leisure-time physical activity.⁵⁴ Hasson et al. demonstrated that integrating aspects of both black-American and white-American cultures was associated with lower diabetes risk (increased pancreatic beta cell compensation), independent of socioeconomic status, physical activity, sedentary time, dietary intake, sex, pubertal development, and fat/fat-free mass.⁵⁵ Despite this evidence suggesting that cultural integration would be inversely associated with obesity risk through health-related behaviors, culture was not associated with BMI in this study. Nevertheless, it is important to note that culturally responsive intervention strategies can aid in the prevention and treatment of obesity in African American youth.⁵⁶

The strengths of this study included the use of accelerometers for measuring MVPA, the FFQ to measure diet, and a comprehensive investigation of social environmental factors using reliable and validated measures. The limitations included a cross-sectional design that relied on self-reported data. High fractions of missing data were noted for daily caloric intake, physical activity, and household education; however, multiple imputation was used to account for this missing data. Finally, the generalizability of this study may be limited to African American youth residing in Southwest Michigan. Despite these limitations, we observed significant relationships between BMI and racial discrimination as well as between BMI and household education.

Current efforts to prevent and treat pediatric obesity in African American youth have primarily focused on behavioral strategies to reduce obesity-related behaviors, including physical inactivity, high-fat/sugar diets, and excessive psychological stress. Less attention has been given to social environmental factors that shape these obesity-related behaviors and obesity outcomes. Although longitudinal studies have been conducted in adults,^14,16,38 these same types of studies are needed in adolescent populations to better understand the mechanisms by which environmental factors increase obesity risk in African American youth. Consistent with previous research, our findings suggest that racial discrimination and household education are important correlates of obesity in African American youth, particularly among African American girls. These findings reinforce the continued need for health education programming and culturally responsive strategies that address the disproportionately high level of environmental stress that African American youth are exposed to. Moreover, behavioral interventions and public policies are needed to better address social and environmental factors associated with obesity risk in ethnic minority pediatric populations.

Footnotes

Acknowledgments

The authors would like to thank the Health & Culture Study team and Michigan Consulting for Statistics, Computing and Analytics Research. They are also grateful to their study participants and their families for their involvement. This work was supported by the University of Michigan Office of Research and the University of Michigan School of Public Health Momentum Center. Clinical Trial Registration: This study is registered at (No. NCT02938663).

Author Disclosure Statement

No competing financial interests exist.

References

Ogden

, Carroll

, Kit

, Flegal

. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 2014; 311:806–814.

Hasson

, Adam

, Davis

, et al. Randomized controlled trial to improve adiposity, inflammation, and insulin resistance in obese African-American and Latino youth. Obesity (Silver Spring), 2012; 20:811–818.

Braveman

, Egerter

, Barclay

. What Shapes Health-Related Behaviors? The Role of Social Factors. Robert Wood Johnson Foundation: Princeton, NJ, 2011.

Ogden

, Lamb

, Carroll

, Flegal

. Obesity and socioeconomic status in children and adolescents: United States, 2005–2008. NCHS Data Brief, 2010:1–8.

Sobal

, Stunkard

. Socioeconomic status and obesity: A review of the literature. Psychol Bull, 1989; 105:260–275.

Gordon-Larsen

, Adair

, Popkin

. The relationship of ethnicity, socioeconomic factors, and overweight in US adolescents. Obes Res, 2003; 11:121–129.

Troiano

, Flegal

. Overweight children and adolescents: Description, epidemiology, and demographics. Pediatrics, 1998; 101(3 Pt 2):497–504.

Boslaugh

, Luke

, Brownson

, et al. Perceptions of neighborhood environment for physical activity: Is it “who you are” or “where you live”?. J Urban Health, 2004; 81:671–681.

Lorenc

, Clayton

, Neary

, et al. Crime, fear of crime, environment, and mental health and wellbeing: Mapping review of theories and causal pathways. Health Place, 2012; 18:757–765.

10.

, Levy

, Peters

, et al. A Walk in the park: The influence of urban parks and community violence on physical activity in Chelsea, MA. Int J Environ Res Public Health, 2016; 13:pii:E97.

11.

Dias

, Whitaker

. Black mothers' perceptions about urban neighborhood safety and outdoor play for their preadolescent daughters. J Health Care Poor Underserved, 2013; 24:206–219.

12.

Brodish

, Cogburn

, Fuller-Rowell

, et al. Perceived racial discrimination as a predictor of health behaviors: The Moderating role of gender. Race Soc Probl, 2011; 3:160–169.

13.

Bermudez-Millan

, Schumann

, Feinn

, et al. Behavioral reactivity to acute stress among Black and White women with type 2 diabetes: The roles of income and racial discrimination. J Health Psychol, 2016; 21:2085–2097.

14.

Cozier

, Wise

, Palmer

, Rosenberg

. Perceived racism in relation to weight change in the Black Women's Health Study. Ann Epidemiol, 2009; 19:379–387.

15.

Vines

, Baird

, Stevens

, et al. Associations of abdominal fat with perceived racism and passive emotional responses to racism in African American women. Am J Public Health, 2007; 97:526–530.

16.

Hunte

, Williams

. The association between perceived discrimination and obesity in a population-based multiracial and multiethnic adult sample. Am J Public Health, 2009; 99:1285–1292.

17.

Berry

JW.

Acculturation as Varieties of Adaptation. Westview Press: Boulder, CO, 1980.

18.

Smedley

Race in North America: Origin and Evolution of a Worldview. Westview Press: Boulder, CO, 1999.

19.

Berry

JW.

Acculturation: Living successfully in two cultures. Int J Intercult Relat, 2005; 29:697–712.

20.

Fanelli Kuczmarski

, Cotugna

, Pohlig

, et al. Snacking and diet quality are associated with the coping strategies used by a socioeconomically diverse urban cohort of African-American and white adults. J Acad Nutr Diet, 2017; 117:1355–1365.

21.

Reynolds

, Killen

, Bryson

, et al. Psychosocial predictors of physical activity in adolescents. Prev Med, 1990; 19:541–551.

22.

Torres

, Nowson

. Relationship between stress, eating behavior, and obesity. Nutrition, 2007; 23:887–894.

23.

Tomiyama

, Puterman

, Epel

, et al. Chronic psychological stress and racial disparities in body mass index change between Black and White girls aged 10–19. Ann Behav Med, 2013; 45:3–12.

24.

Rhee

, Sampson

, Cho

, et al. Comparison of methods to account for implausible reporting of energy intake in epidemiologic studies. Am J Epidemiol, 2015; 181:225–233.

25.

Johnson

RK.

Dietary intake—How do we measure what people are really eating?. Obes Res, 2002; 10(Suppl 1):63S–68S.

26.

Huang

, Howarth

, Lin

, et al. Energy intake and meal portions: Associations with BMI percentile in U.S. children. Obes Res, 2004; 12:1875–1885.

27.

Suglia

, Ryan

, Wright

. Creation of a community violence exposure scale: Accounting for what, who, where, and how often. J Trauma Stress, 2008; 21:479–486.

28.

Fisher

, Wallace

, Fenton

. Discrimination distress during adolescence. J Youth Adolesc, 2000; 29:679–695.

29.

Unger

, Gallaher

, Shakib

, et al. The AHIMSA acculturation scale: A new measure of acculturation for adolescents in a multicultural society. J Early Adolesc, 2002; 22:225–251.

30.

Shorr

IJ.

How to Weigh and Measure Children. United Nations: New York, 1986; revised Shorr IJ, 1998.

31.

James-Todd

, Tehranifar

, Rich-Edwards

, et al. The impact of socioeconomic status across early life on age at menarche among a racially diverse population of girls. Ann Epidemiol, 2010; 20:836–842.

32.

Carskadon

, Acebo

. A self-administered rating scale for pubertal development. J Adolesc Health, 1993; 14:190–195.

33.

Petersen

, Crockett

, Richards

, Boxer

. A self-report measure of pubertal status: Reliability, validity, and initial norms. J Youth Adolesc, 1988; 17:117–133.

34.

Cain

, Sallis

, Conway

, et al. Using accelerometers in youth physical activity studies: A review of methods. J Phys Act Health, 2013; 10:437–450.

35.

Puyau

, Adolph

, Vohra

, Butte

. Validation and calibration of physical activity monitors in children. Obes Res, 2002; 10:150–157.

36.

Araujo

, Yokoo

, Pereira

. Validation and calibration of a semiquantitative food frequency questionnaire designed for adolescents. J Am Diet Assoc, 2010; 110:1170–1177.

37.

Cohen

, Kamarck

, Mermelstein

. A global measure of perceived stress. J Health Soc Behav, 1983; 24:385–396.

38.

Cunningham

, Berkman

, Kawachi

, et al. Changes in waist circumference and body mass index in the US CARDIA cohort: Fixed-effects associations with self-reported experiences of racial/ethnic discrimination. J Biosoc Sci, 2013; 45:267–278.

39.

Chang

, Hillier

, Mehta

. Neighborhood racial isolation, disorder and obesity. Soc Forces, 2009; 87:2063–2092.

40.

LeBlanc

, Janssen

. Difference between self-reported and accelerometer measured moderate-to-vigorous physical activity in youth. Pediatr Exerc Sci, 2010; 22:523–534.

41.

Ervin

, Ogden

. Trends in intake of energy and macronutrients in children and adolescents from 1999–2000 through 2009–2010. NCHS Data Brief, 2013:1–8.

42.

Gee

, Ro

, Gavin

, Takeuchi

. Disentangling the effects of racial and weight discrimination on body mass index and obesity among Asian Americans. Am J Public Health, 2008; 98:493–500.

43.

Chambers

, Tull

, Fraser

, et al. The relationship of internalized racism to body fat distribution and insulin resistance among African adolescent youth. J Natl Med Assoc, 2004; 96:1594–1598.

44.

Wingenfeld

, Kuehl

, Boeker

, et al. Stress reactivity and its effects on subsequent food intake in depressed and healthy women with and without adverse childhood experiences. Psychoneuroendocrinology, 2017; 80:122–130.

45.

Lewis

, Kravitz

, Janssen

, Powell

. Self-reported experiences of discrimination and visceral fat in middle-aged African-American and Caucasian women. Am J Epidemiol, 2011; 173:1223–1231.

46.

Shelton

, Puleo

, Bennett

, et al. The association between racial and gender discrimination and body mass index among residents living in lower-income housing. Ethn Dis, 2009; 19:251–257.

47.

Johnson

, Risica

, Gans

, et al. Association of perceived racial discrimination with eating behaviors and obesity among participants of the SisterTalk study. J Natl Black Nurses Assoc, 2012; 23:34–40.

48.

Corral

, Landrine

, Hao

, et al. Residential segregation, health behavior and overweight/obesity among a national sample of African American adults. J Health Psychol, 2012; 17:371–378.

49.

Butler

, Tull

, Chambers

, Taylor

. Internalized racism, body fat distribution, and abnormal fasting glucose among African-Caribbean women in Dominica, West Indies. J Natl Med Assoc, 2002; 94:143–148.

50.

Tull

, Wickramasuriya

, Taylor

, et al. Relationship of internalized racism to abdominal obesity and blood pressure in Afro-Caribbean women. J Natl Med Assoc, 1999; 91:447–452.

51.

Gooding

, Milliren

, Austin

, et al. Exposure to violence in childhood is associated with higher body mass index in adolescence. Child Abuse Negl, 2015; 50:151–158.

52.

Dubois

WEB.

Souls of Black Folk. A.C. McClurg: Chicago, IL, 1903.

53.

Dressler

, Bindon

, Neggers

. Culture, socioeconomic status, and coronary heart disease risk factors in an African American community. J Behav Med, 1998; 21:527–544.

54.

Airhihenbuwa

, Kumanyika

, TenHave

, Morssink

. Cultural identity and health lifestyles among African Americans: A new direction for health intervention research?. Ethn Dis, 2000; 10:148–164.

55.

Hasson

, Adam

, Pearson

, et al. Sociocultural and socioeconomic influences on type 2 diabetes risk in overweight/obese African-American and Latino-American children and adolescents. J Obes, 2013; 2013:512914.

56.

McClelland

, Weekes

, Bazzi

, et al. Perception of obesity in African-American and Arab-American minority groups. J Racial Ethn Health Disparities, 2016; 3:160–167.