Abstract
Background:
Prevalence of certain disordered eating behaviors is higher among Hispanic youth compared to non-Hispanics. Understanding the role of body image and social attitudes towards weight in disordered eating may inform treatment in Hispanic youth.
Methods:
We analyzed data from the Hispanic Community Health Study/Study of Latino Youth (SOL Youth). Our sample included 1,463 children aged 8–16 years from four sites (Bronx, Chicago, Miami, San Diego) assessed in 2011–2014. Body image discrepancy score was calculated as the difference between perceived ideal body image and actual body image using two numbered visual graphs: adolescent (n = 728) or child (n = 735), each with slightly different scales. Questionnaires measured influences from social attitudes toward weight and disordered eating behaviors. Three disordered eating behaviors (dieting, overeating, and compensatory behaviors) were analyzed as the dependent variable. Logistic regression models adjusted for age, sex, acculturative stress, and field center to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CI).
Results:
Mean body image discrepancy score was −0.79 for adolescents (SE = 0.08) and −0.50 for children (SE = 0.05), with a negative score signifying a perceived actual body image larger than their ideal. Body image discrepancy was strongly associated with dieting (dieting ≥5 times/year aOR = 0.64, 95% CI 0.53, 0.77) and compensatory behaviors (aOR = 0.65, 95% CI 0.50, 0.85) among adolescents, and was strongly associated with overeating among children (aOR = 0.74, 95% CI 0.61, 0.91). Significant associations were not observed with social attitudes towards weight.
Conclusions:
Associations observed with body image discrepancy and disordered eating behaviors can inform interventions in Hispanic/Latino youth, which should consider acculturative stress.
Keywords: Hispanic youth, Disordered eating, Body image, Social influences, Overeating, Body mass index, Acculturative stress
1. Introduction
Disordered eating behaviors are more common among Hispanic/Latino adolescents compared to their non-Hispanic/Latino counterparts (Croll, Neumark-Sztainer, Story, & Ireland, 2002; Swanson, Crow, Le Grange, Swendsen, & Merikangas, 2011). Disordered eating behaviors include binge eating and compulsive eating behaviors, dieting and restrained eating, or compensatory behaviors. Body image dissatisfaction is a known factor associated with disordered eating (Croll et al., 2002; Neumark-Sztainer, Paxton, Hannan, Haines, & Story, 2006) and can be more common in teenagers with excess weight (Neumark-Sztainer, Story, Hannan, Perry, & Irving, 2002). While obesity is a significant health problem in the United States, it is more common among Hispanic/Latino youth (Hales, Carroll, Fryar, & Ogden, 2017). Our objective in this analysis is to determine if body image, and sociocultural ideals that may influence body image, have associations with disordered eating behaviors in Hispanic/Latino youth, and to examine if these associations persist after adjustment for the effects of acculturation.
Self-discrepancy theory suggests that body image discrepancy, or the discrepancy between what an individual perceives as their ideal body shape and their actual body shape, can have emotional consequences that can lead an individual to engage in behaviors to reduce the discrepancy (Vartanian, 2012). In one study among Hispanic/Latino youths, body image discrepancy is associated with being overweight (Ayala, Mickens, Galindo, & Elder, 2007). However, body image discrepancy can be measured in values with opposing directions, one in which the individual perceives the ideal larger than the actual, and the other in which the individual perceives the ideal thinner than the actual, both which can result in dissatisfaction. Hispanic females are more likely to perceive a body size that is larger than the ideal compared to non-Hispanic White females (Neumark-Sztainer, Croll, et al., 2002; Robinson et al., 1996). Latino males, however, are less likely to report a desire to be thinner compared to Latino females (Ayala et al., 2007). In one study of adolescent girls, body image discrepancy was predictive of disordered eating in both non-Hispanic White and Hispanic/bi-ethnic Hispanic girls (Erickson & Gerstle, 2007). In one study, Hispanic women reported greater body image discrepancy at lower body mass index (BMI) levels compared to Black and White women (Fitzgibbon, Blackman, & Avellone, 2000). Examining moderation by sex is therefore necessary as females may have greater negative body image discrepancy and increased disordered eating.
Social norms and cultural trends, particularly in the media, can influence how youths perceive ideals in relation to weight, beauty, and body size and shape (Forney & Ward, 2013; Grabe, Ward, & Hyde, 2008; Labre, 2002; Pike, Hoek, & Dunne, 2014). Internalization of these sociocultural ideals can lead to dissatisfaction and disordered eating behaviors, which can increase the risk for eating disorders (Vander Wal, Gibbons, & Grazioso, 2008). These influences can differ by gender, where females may feel pressure to be thinner (Grabe et al., 2008), while the male body ideal is to appear more muscular (Labre, 2002). Differences are also noted by race/ethnicity, with Black women having less awareness of the societal appearance norms set by the media than other racial/ethnic groups (Quick & Byrd-Bredbenner, 2014). However, there does not appear to be a consensus in the literature on how Hispanic/Latino youth are influenced by the media and sociocultural norms. While some studies report that Hispanic/Latina girls may be vulnerable to sociocultural influences (Austin & Smith, 2008; Snapp, 2009), in another study, media-influenced sociocultural values and ideals do not contribute to disordered eating among Latino youth (Lopez, Corona, & Halfond, 2013). These conflicting findings may be due to Hispanic/Latina women having sociocultural influences of both the Latina curvy ideal and the American thin ideal (Franko et al., 2012), which may range widely based on different acculturation levels and across subpopulations of Hispanic/Latina women.
Research on the role of body image and social attitudes towards weight ideals in association with disordered eating behaviors is lacking for Hispanic/Latino youth, particularly in large samples with diverse ethnic background groups. Differences in acculturation across Hispanic/Latino background groups could confound this relationship, as acculturation can influence both body image ideals and sociocultural influences. However, the existing research on this is conflicting. One early study showed that as Hispanic/Latina girls become more assimilated into the dominant Anglo culture, body image dissatisfaction and disordered eating behaviors increased (Gowen, Hayward, Killen, Robinson, & Taylor, 1999). However, research by Ayala et al. (2007) found conflicting associations between acculturation and body image dissatisfaction. It may be that a more appropriate measure is acculturative stress, or the psychological effect of adapting to a new culture, which better captures the negative impact of acculturation on Hispanic/Latino youth (Smart & Smart, 1995). This inconsistency in the literature indicates a need for further consideration of the role of acculturation as well as appropriate measures for capturing its effect on health.
Large epidemiologic studies are needed to understand how body image discrepancy and sociocultural attitudes towards weight are associated with disordered eating in Hispanic/Latino youths in the United States. Previous studies conducted with small clinical samples or relatively small community-based samples have not been able to accurately assess prevalence among youths of diverse Hispanic/Latino background groups. Most of the literature on disordered eating in Hispanic/Latinos also focuses on adolescents and young adults, with few findings available on children. Our study aims to address these gaps in the literature to inform future studies on disordered eating in Hispanic/Latino youths. Therefore, we sought to examine body image discrepancy and sociocultural attitudes towards weight in relation to disordered eating in Hispanic/Latino youth, considering age group and sex. We hypothesized that negative body image discrepancy (i.e., perceived actual larger than ideal) and endorsement of sociocultural attitudes toward a thin ideal would be associated with higher odds of disordered eating behavior.
2. Methods
2.1. Study population and procedures
The Hispanic Community Health Study/Study of Latinos (HCHS/SOL) is a population-based cohort study of chronic diseases among adults who identify as Hispanic/Latino. HCHS/SOL participants (n = 16415) aged 18–74 were recruited using 2-stage probability sampling during 2008–2011 from four U.S. communities (Chicago, IL; Miami, FL; Bronx, NY; San Diego, CA). Details regarding the sampling design (LaVange, 2010) and protocol (Sorlie, 2010) have been reported previously.
The Hispanic Community Health Study/Study of Latino Youth (SOL Youth) is a cross-sectional study of youth ages 8–16 years, living with a HCHS/SOL participant, and free from known serious health issues that could interfere with completing the examination. All youth must have been able to read, listen, and write in English or Spanish. The SOL Youth study is described in more detail elsewhere (Isasi et al., 2014). All participants underwent a clinical examination and interview by trained staff at one of the four field centers. Clinical examination included the collection of anthropometric measures along with phlebotomy and other health exams. Extensive data was collected at interview, including demographics, dietary information, physical activity, psychological health, family functioning, and social and cultural beliefs. SOL Youth was conducted with approval from institutional review boards at each center. Written informed consent and assent were obtained from all parent/caregivers and children, respectively.
This analysis focuses on measures collected from the SOL Youth study. Of 1777 eligible youth identified through screening, 1466 (82%) participated between 2012 and 2014. Our sample excludes youth with missing data on all independent and dependent variables (n = 3), leaving a final analytic sample of 1463. The independent variables are body image discrepancy score and the sum score on the social attitudes towards weight questionnaire (discussed in more detail below). The dependent variable is disordered eating behaviors. As disordered eating is associated with being overweight or obese in Latino youth (Ayala et al., 2007), we also include BMI as a comparison outcome.
2.2. Measures
2.2.1. Body image
Following validated procedures by Ayala et al. (2007), SOL Youth participants were shown either one of two body image cards with drawings of male and female bodies ranging in size (Allison, 1995). The card shown depended on the interviewer’s perception of whether the participant was a child or adolescent, with no set age for determining which set of images to use. The adolescent card showed images of increasing body sizes, ranging from 1 (smallest) to 9 (largest), and the child card showed images of bodies ranging from 1 (smallest) to 7 (largest). The participant was instructed to indicate which body aligned with their perception of their actual body size and which aligned with their perception of an ideal body size. Body image discrepancy score was calculated as the difference between the perceived ideal body size and actual body size, with a negative score signifying an actual body size larger than their ideal and a positive score signifying an actual body size smaller than their ideal. Additional analyses were also conducted using an absolute value of the discrepancy score to determine if the effect was the same regardless of the direction of the discrepancy. We further restricted these models to those with positive (≥0) and negative (≤0) body image discrepancy scores to examine the differences in direction.
2.2.2. Social attitudes towards weight
SOL Youth participants were asked about social attitudes towards weight in a brief 7-item questionnaire, which is a modified version of a more extensive questionnaire, the Social Attitudes Towards Appearance Questionnaire (SATAQ) (Heinberg & Thompson, 1995). The SATAQ has been validated for use among different ethnic groups, including Hispanic/Latinos (Warren, Gleaves, & Rakhkovskaya, 2013). The modified questionnaire measures the youth’s endorsement of social ideals on appearance and the role of media influences on body image by including questions from the internalization subscale of the SATAQ. Responses were on a five-point Likert type scale, ranging from ‘Completely Disagree’ (1) to ‘Completely Agree’ (5). The only changes made to the items from this subscale were to adapt them to be gender-neutral. Examples of questions include “I believe that clothes look better on thin models” and “I tend to compare my body to people in magazines and TV”, where “people” replaced “women” from the original item in the SATAQ. Responses were summed into a total score, in which a higher score reflects greater social influences on a youth’s attitudes towards weight.
2.2.3. Disordered eating behaviors and BMI percentile
The SOL Youth disordered eating questionnaire was adapted from the Minnesota Adolescent Health Survey (Neumark-Sztainer, Story, Resnick, & Blum, 1998) and was comprised of questions concerning dieting and disordered eating behaviors. The 4-item questionnaire focused on three areas: 1) how often they diet (0 = never to 4 = dieting ≥5 times a year), 2) overeating or feelings of being unable to control the amount they eat (yes/no response, asked in two different questions and combined), and 3) compensatory behaviors, such as vomiting or taking ipecac, laxatives, or diuretics (yes/no response, asked separately). Three outcomes of diet frequency (categorical), any overeating (dichotomous), and any compensatory behaviors (dichotomous) were created.
Weight was measured for youth on a digital scale (Tanita Body Composition Analyzer, TBF 300, Japan) to the nearest 0.1 kg and height to the nearest cm using a wall-mounted stadiometer (SECA 222, Germany) using a standardized protocol across sites. Body mass index (BMI) in kg/m2 was converted to BMI-for-age percentiles using sex-specific scales from the Centers for Disease Control and Prevention. BMI percentile was used as a comparison outcome variable, assuming that if body image discrepancy and social attitudes towards weight affect disordered eating behavior, this would also be reflected in association with BMI percentile.
2.2.4. Covariates
Known confounders such as youth age and sex were included as covariates in our analysis. To account for the influence of other sociocultural factors, we considered other correlates in our statistical models such as Hispanic/Latino background, study site, acculturation, and acculturative stress. Demographic factors (age, sex, Hispanic/Latino background) were self-reported at time of interview in the youth’s preferred language of English or Spanish. Hispanic/Latino background included identifying as (or identifying as a descendent of) Mexican, Central American, South American, Cuban, Puerto Rican, Dominican, or other/more than one background. Age was measured from the date of birth reported at time of interview. The interview was conducted at one of four study sites across the United States in Chicago, Illinois, Miami, Florida, Bronx, New York, and San Diego, California.
This is the first study to examine differences in disordered eating in a diverse sample of Hispanic/Latino backgrounds. We expect for differences in disordered eating, body image discrepancy, and BMI to emerge by background, consistent with other studies in the HCHS/SOL. Acculturation was based on the Acculturation, Habits, and Interests Multicultural Scale for Adolescents (AHIMSA), which assessed cultural preferences and beliefs among youths (Unger et al., 2002). Based on the scoring protocols developed by Unger, Ritt-Olson, Wagner, Soto, and Baezconde-Garbanati (2007), we used eight dummy variables to measure the four acculturation groups theorized by Barry (2005) with all items summed for a score from 0 to 8 for variables measuring assimilation (α = 0.64), integration (α = 0.68), separation (α = 0.57), and marginalization (α = 0.42). A categorical variable identifying acculturation groups was defined by categorizing youth with assimilation scores ≥4.0 as assimilated; youth with integrated scores ≥4.0 as integrated; youth with separated scores ≥4.0 as separated; and youth with marginalized scores ≥4.0 as marginalized. The marginalized and separated categories were combined since relatively few youth fell into the marginalized category (N = 2). In addition to AHIMSA, the Acculturative Stress Index measured language conflict, intra-family cultural conflict, and discrimination (Gil & Vega, 1996). All items were averaged for an acculturative stress score ranging from 1 to 5, with a 5 representing the highest level of acculturative stress, with the scale having good reliability (α = 0.73) in our sample (Perreira et al., 2019).
2.3. Statistical analysis
To account for the HCHS/SOL complex survey design, all analyses incorporated SOL Youth weights for sampling, stratification, and clustering. All analyses were also stratified according to the body image card the participant received, either child or adolescent, due to the different scales. Descriptive statistics were obtained for demographics, independent variables, dependent variables, and covariates. Body image discrepancy scores were compared by sex to determine if differences were present using t-tests and an alpha level of α = 0.05. Alpha levels and hypothesis were set a priori Frequencies and means of dependent and independent variables were obtained by Hispanic/Latino background.
For adolescents and children, separately, dieting frequency, overeating, and compensatory behaviors were regressed on each independent variable, body image discrepancy and social attitudes towards weight sum score. Logistic regression models were run unadjusted and adjusted for youth age, sex, study site, and acculturative stress, to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CI). Confounders were originally selected using a directed acyclic graph approach to identify a minimally sufficient set of covariates (VanderWeele & Robins, 2007). To further reduce covariates that may be collinear, we included age and sex in all models and conducted a forward stepwise regression adding the covariates in the model for Hispanic/Latino background, AHIMSA category for acculturation, study site, and acculturative stress and used results yielding a p value of <0.05 as the criterion for inclusion. Study site and acculturative stress were included in our final adjustment set. Hispanic/Latino background and AHIMSA category did not significantly change our estimates. We considered moderation by sex by separately including the variable as an interaction term in the model, but removed the interaction term if not significant at a p-value of p < 0.05.
A sensitivity analysis was conducted for the regression models to test the effect of potential body image card misclassification due to age. Using frequencies of assignment by age, a mid-point age was identified at age 12 years in which a similar number of youth were assigned a child card or an adolescent card. Of youth assigned an adolescent card, 72.6% were ≥12 years of age and 70.2% of youth assigned a child card were ≤12 years of age. Regression models were then run restricting each body image card to these cut-points for age. Separate linear regression models were also run with BMI-for-age percentile as the outcome variable in order to examine if the association also extended to youth’s weight. We also considered the possible interaction with body image discrepancy and BMI, and therefore tested for interaction by including an interaction term in the models and removing the term if it was not significant (p < 0.05). Additional analyses were also conducted using absolute values of body image discrepancy score. Analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC) and Mplus 8 (Muthén & Muthén, Los Angeles, CA). The analytic plan for this analysis was pre-specified and approved by the HCHS/SOL Publications Committee before access to the data was obtained.
3. Results
Out of 1,463 Hispanic/Latino youth, 728 were given an adolescent body image card and 735 were given a child body image card. As shown in Table 1, ages ranged from 8 to 16 years in both groups, with a mean age of 13 years for adolescents and 11 years for children. The adolescent group had a high proportion of females (57.2%) and a mean body image discrepancy score of −0.79 (SE = 0.08; range = −7, 6). The child group had a low proportion of females (42.8%) and a mean body image discrepancy of −0.50 (SE = 0.05, range = −5, 3). The mean actual body size was larger than the mean ideal body size for both the child and adolescent groups. No difference in body image discrepancy was detected by sex among children (t = −0.63, p = 0.53). However, statistically significant differences by sex were seen among adolescents, with a higher discrepancy among females (t = −2.35, p = 0.02). Similarly, no difference in social attitudes towards weight score was detected by sex among children (t = 1.71, p = 0.09), but significant differences were noted among adolescents (t = 2.25, p = 0.03). The acculturative stress index score was similar among children (mean = 1.67; SE = 0.03) and adolescents (mean = 1.55, SE = 0.03). Mean percentiles for BMI were similar among children and adolescents. Approximately half of both adolescents (50.3%) and children (50.2%) reported never dieting; however, a higher proportion of children than adolescents (19.3% vs. 15.7%) reported dieting five or more times a year (Table 1). A higher proportion of children than adolescents also reported overeating (30.4% vs. 24.7%) and compensatory behaviors (4.3% vs. 3.2%).
Table 1.
SOL Youth characteristics (N = 1463).
| Total Sample (1463) |
Adolescent Card (N = 735) |
Child Card (N = 728) |
||||
|---|---|---|---|---|---|---|
| Unweighted N |
Weighted % |
Unweighted N |
Weighted % |
Unweighted N |
Weighted % |
|
| Female | 735 | 48.7 | 389 | 57.2 | 346 | 42.8 |
| Hispanic/Latino Background | ||||||
| Central American | 112 | 6.3 | 64 | 7.2 | 48 | 5.3 |
| Cuban | 103 | 5.7 | 47 | 5.3 | 56 | 6.1 |
| Dominican | 167 | 13.4 | 99 | 14.7 | 68 | 11.9 |
| Mexican | 647 | 48.6 | 324 | 45.8 | 323 | 52.0 |
| Puerto Rican | 128 | 9.9 | 67 | 10.7 | 61 | 8.9 |
| South American | 67 | 4.1 | 34 | 4.0 | 33 | 4.2 |
| Mixed/Other | 134 | 10.0 | 73 | 10.2 | 61 | 9.8 |
| Center | ||||||
| Bronx | 422 | 35.9 | 231 | 37.6 | 191 | 34.0 |
| Chicago | 370 | 16.1 | 186 | 16.0 | 184 | 16.2 |
| Miami | 262 | 13.6 | 133 | 13.6 | 129 | 13.6 |
| San Diego | 409 | 34.4 | 191 | 32.8 | 218 | 36.2 |
| Dieting Frequency | ||||||
| Never | 713 | 50.3 | 374 | 50.3 | 339 | 50.2 |
| 1-4 times/year | 476 | 32.3 | 249 | 34.0 | 227 | 30.4 |
| ≥5 times/year | 262 | 17.4 | 115 | 15.7 | 147 | 19.3 |
| Overeating | 409 | 27.3 | 182 | 24.7 | 227 | 30.4 |
| Compensatory behaviors |
61 |
3.7 |
27 |
3.2 |
34 |
4.3 |
| Weighted Mean | SE | Weighted Mean | SE | Weighted Mean | SE | |
| Body Image Discrepancy | – | – | −0.79 | 0.08 | −0.50 | 0.05 |
| Actual body size | 5.44 | 0.08 | 4.27 | 0.04 | ||
| Ideal body size | 4.64 | 0.06 | 3.77 | 0.05 | ||
| Social Attitudes Towards Weight Score | 17.43 | 0.22 | 17.28 | 0.30 | 17.61 | 0.32 |
| Acculturative Stress | 1.61 | 0.02 | 1.55 | 0.03 | 1.67 | 0.03 |
| BMI Percentile | 72.5 | 1.0 | 73.5 | 1.3 | 71.4 | 1.5 |
The distribution of disordered eating behaviors varied by Hispanic/Latino background (Table 2). Among adolescents, those of Central American background had the highest proportion of not having dieted (59.9%) and South Americans had the highest proportion reporting dieting ≥5 times/year (28.7%). Cuban adolescents were the only group to have a null mean body image discrepancy. However, this coincided with a high mean sum score on the influence of social attitudes towards weight and low mean BMI percentile. Among children, Mexicans had the highest proportion of not having dieted (53.9%) while Cuban children had the highest proportion reporting dieting ≥5 times/year (38.0%). Children of Central American background had a high proportion of overeating (41.9%) and low mean body image discrepancy (mean = −0.8). Children of Dominican background had the highest mean BMI percentile. Sample sizes were small for some Hispanic/Latino background groups and those reporting compensatory behaviors. Therefore, these comparisons were underpowered to make statistically significant inferences.
Table 2.
Mean scores (SE) and frequencies (%) of dependent and independent variables by Hispanic Latino/Background.
| Adolescents | Dominican |
Puerto Rican |
Cuban |
Central American |
Mexican |
South American |
Mixed/Other |
|---|---|---|---|---|---|---|---|
| N = 99 |
N = 67 |
N = 47 |
N = 64 |
N = 324 |
N = 34 |
N = 73 |
|
| weighted % | weighted % | weighted % | weighted % | weighted % | weighted % | weighted % | |
| Dieting Frequency | |||||||
| Never | 56.3 | 42.7 | 52.5 | 59.9 | 51.2 | 47.4 | 45.8 |
| 1-4 times/year | 35.7 | 36.9 | 38.1 | 28.1 | 31.2 | 23.9 | 43.4 |
| ≥5 times/year | 8.0 | 20.4 | 9.4 | 9.4 | 17.5 | 28.7 | 10.8 |
| Overeating | 18.9 | 23.8 | 17.9 | 24.8 | 27.5 | 24.6 | 45.7 |
| Compensatory behaviors |
3.2 |
3.6 |
0 |
4.5 |
2.8 |
0 |
2.3 |
| Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | |
| Body Image Discrepancy | −0.8 (0.2) | −1.1 (0.3) | 0.0 (0.3) | −0.7 (0.2) | −0.7 (0.1) | −1.0 (0.3) | −1.1 (0.2) |
| Social Attitudes Towards Weight Sum Score | 16.4 (0.8) | 15.5 (1.0) | 18.8 (1.2) | 18.2 (1.2) | 17.5 (0.5) | 19.4 (1.2) | 17.0 (0.8) |
| BMI Percentile |
76.8 (2.8) |
76.8 (4.1) |
60.4 (5.8) |
65.7 (5.8) |
73.2 (1.9) |
72.3 (5.1) |
77.72 (3.4) |
| Children | N = 68 | N = 61 | N = 56 | N = 48 | N = 323 | N = 33 | N = 61 |
| weighted % | weighted % | weighted % | weighted % | weighted % | weighted % | weighted % | |
| Dieting Frequency | |||||||
| Never | 51.9 | 49.4 | 38.2 | 46.0 | 53.9 | 32.5 | 55.4 |
| 1-4 times/year | 33.9 | 35.4 | 23.9 | 24.2 | 29.1 | 40.7 | 26.0 |
| ≥5 times/year | 14.2 | 15.2 | 38.0 | 29.8 | 17.0 | 26.9 | 18.6 |
| Overeating | 27.4 | 20.0 | 31.5 | 41.9 | 30.7 | 39.3 | 34.7 |
| Compensatory behaviors |
9.6 |
8.7 |
8.8 |
8.7 |
2.1 |
3.2 |
0.2 |
| Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | Mean (SE) | |
| Body Image Discrepancy | −0.6 (0.2) | −0.4 (0.2) | −0.6 (0.1) | −0.8 (0.2) | −0.5 (0.1) | −0.6 (0.2) | −0.6 (0.2) |
| Social Attitudes Towards Weight Sum Score | 17.9 (1.2) | 18.4 (1.0) | 18.7 (1.1) | 17.9 (1.1) | 17.1 (0.5) | 18.3 (1.6) | 17.7 (0.9) |
| BMI Percentile | 76.4 (4.2) | 66.2 (5.2) | 72.7 (4.2) | 69.4 (5.4) | 70.1 (2.3) | 74.1 (3.2) | 73.2 (5.0) |
3.1. Body image discrepancy and disordered eating
Our results show an association between body image discrepancy and disordered eating. Among adolescents, body image discrepancy was negatively associated with dieting, overeating, and compensatory behaviors, indicating higher odds of disordered eating with negative body image score (Table 3). These associations remained after adjustment, with body image discrepancy associated with dieting 1–4 times a year (aOR = 0.75, 95% CI 0.66, 0.84) and with dieting ≥5 times a year (aOR = 0.65, 95% CI 0.53, 0.77) among adolescents. Associations were also observed with body image discrepancy and overeating (aOR = 0.86, 95% CI 0.76, 0.98) and compensatory behaviors (aOR = 0.65, 95% CI 0.50, 0.84). Acculturation was not a significant predictor in these models, however acculturative stress was strongly associated with dieting ≥5 times a year (aOR = 1.82, 95% CI 1.05, 3.15), overeating (aOR = 2.04, 95% CI 1.38, 3.03), and compensatory behaviors (aOR = 2.09, 95% CI 1.03, 4.11). Among children, body image discrepancy was negatively associated with dieting (1–4 times/year a OR = 0.74, 95% CI 0.61, 0.91; ≥5 times/year aOR = 0.74, 95% CI 0.61, 0.91). However, no significant association was observed with body image discrepancy and compensatory behaviors (aOR = 1.05, 95% CI 0.75, 1.47). Similar as with adolescents, in adjusted models acculturative stress was strongly associated with dieting ≥5 times a year (aOR = 1.69, 95% CI 1.15, 2.49), overeating (aOR = 2.51, 95% CI 1.71, 3.68), and compensatory behaviors (aOR = 2.02, 95% CI 1.16, 3.54). Moderation with sex was not observed in any of the regression models. Additionally, Hispanic/Latino background and AHIMSA category for acculturation did not significantly change our estimates and were not included in the final adjustment set. Acculturative stress in all models as it was significantly associated with all disordered eating behaviors.
Table 3.
Logistic regression analyses for dieting, overeating, compensatory behaviors.
| Dieting | Overeating | Purging | ||||||
|---|---|---|---|---|---|---|---|---|
| 1-4 times a year |
≥5 times a year |
|
|
|||||
| OR (95% CI) | Adjusted ORa (95%CI) |
OR (95% CI) | Adjusted ORa (95%CI) |
OR (95% CI) | Adjusted ORa (95% CI) |
OR (95% CI) | Adjusted ORa (95% CI) |
|
| Adolescents | ||||||||
| Body Image Discrepancy | 0.75 (0.67, 0.84) | 0.75 (0.66, 0.84) | 0.66 (0.56, 0.79) | 0.64 (0.53, 0.77) | 0.88 (0.79, 0.99) | 0.86 (0.76, 0.98) | 0.71 (0.56, 0.91) | 0.65 (0.50, 0.84) |
| Social Attitudes Towards Weight Score | 1.03 (1.00, 1.06) | 1.03 (1.00, 1.06) | 1.05 (1.02, 1.09) | 1.06 (1.01, 1.11) | 1.02 (0.99, 1.05) | 1.01 (0.98, 1.05) | 0.97 (0.91, 1.04 | 0.96 (0.90, 1.03) |
| Children | ||||||||
| Body Image Discrepancy | 0.78 (0.64, 0.96) | 0.74 (0.61, 0.91) | 0.76 (0.63, 0.93) | 0.74 (0.61, 0.91) | 0.75 (0.62, 0.91) | 0.74 (0.61, 0.91) | 1.01 (0.70, 1.47) | 1.05 (0.75, 1.47) |
| Social Attitudes Towards Weight Score | 1.03 (0.99, 1.06) | 1.02 (0.99, 1.06) | 1.03 (0.99, 1.07) | 1.01 (0.98, 1.05) | 1.04 (1.01, 1.08) | 1.03 (1.00, 1.07) | 1.03 (0.97, 1.08) | 1.00 (0.94, 1.06) |
Adjusted for youth age, sex, study site, acculturative stress.
Further analyses using absolute values of body image discrepancy scores show a slight attenuation of effect measures towards the null (Table 4). When examining associations between those with negative versus positive scores, associations were stronger among youth with body image discrepancy scores in the negative range compared to those with body image discrepancy scores in the positive range. Analyses examining interaction between body image discrepancy and BMI percentiles were not significant for all dependent variables. However, analyses examining associations with BMI show that a negative body image discrepancy was associated with BMI percentile among both adolescents (Table 5; adjusted β = −4.63, 95% CI −5.83, −3.43) and children (adjusted β = −6.82, 95% CI −9.33, −4.11).
Table 4.
Logistic regression analyses for the association between absolute value of body image discrepancy score and dieting, overeating, compensatory behaviors.
| Dieting | Overeating | Compensatory Behaviors |
||
|---|---|---|---|---|
|
|
|
|
||
| 1-4 times a year |
≥5 times a year |
|||
|
|
|
|
|
|
| Adjusted ORa (95%CI) |
Adjusted ORa (95%CI) |
Adjusted ORa (95% CI) |
Adjusted ORa (95% CI) |
|
| Absolute value, overall | ||||
| Adolescent | 1.13 (0.95, 1.34) | 1.53 (1.21, 1.94) | 1.03 (0.86, 1.25) | 1.46 (1.01, 2.10) |
| Child | 1.36 (1.06, 1.75) | 1.31 (0.97, 1.77) | 1.50 (1.18, 1.89) | 0.99 (0.75, 1.31) |
| Among negative discrepancy | ||||
| Adolescent | 1.23 (1.01, 1.48) | 1.69 (1.32, 2.17) | 1.02 (0.85, 1.23) | 1.41 (1.01, 1.98) |
| Child | 1.41 (1.07, 1.84) | 1.37 (1.01, 1.85) | 1.52 (1.19, 1.95) | 0.95 (0.71, 1.26) |
| Among positive discrepancy | ||||
| Adolescent | 0.55 (0.36, 0.85) | 0.91, (0.57, 1.05) | 0.58 (0.33, 1.01) | 0.15 (0.02, 1.08) |
| Child | 1.01 (0.64, 1.60) | 0.92 (0.50, 1.67) | 1.19 (0.68, 2.07) | 1.80 (0.83, 3.89) |
Adjusted for youth age, sex, study site, acculturative stress.
Table 5.
Results from linear regression analyses for BMI percentile.
| Adolescents | BMI Percentile | |
|---|---|---|
| β (95% CI) | Adjusted βa(95% CI) | |
| Body Image Discrepancy | −4.60 (−5.78, −3.42) | −4.63 (−5.83, −3.43) |
| Social Attitudes Towards Weight Score | 0.44 (0.09, 0.80) | 0.63 (0.28, 0.98) |
| Children | ||
| Body Image Discrepancy | −6.78 (−9.35, −4.21) | −6.82 (−9.33, −4.11) |
| Social Attitudes Towards Weight Score | 0.17 (−0.25, 0.60) | 0.12 (−0.31, 0.56) |
Adjusted for youth sex, age, study site, acculturative stress.
In sensitivity analyses, a slightly stronger negative association between body image discrepancy and dieting was observed for both children and adolescents when ages were restricted to ≤12 and ≥12 years, respectively. However, no substantial changes were observed with overeating or compensatory behaviors, and confidence intervals were wider due to the smaller sample size (see Supplemental Table).
3. 2. social attitudes towards weight and disordered eating
No strong associations were observed with social attitudes towards weight and dieting, overeating, or compensatory behaviors among adolescents or children. Slightly significant associations were observed with social attitudes towards weight and dieting (<5 times aOR = 1.06 (95% CI 1.01, 1.11) among adolescents. In adolescent models, acculturative stress was positively associated with dieting ≥5 times a year (aOR = 1.50, 95% CI 0.84, 2.67), overeating (aOR = 1.95, 95% CI 1.33, 2.86), and compensatory behaviors (aOR = 2.06 (95% CI 1.03, 4.12). A slightly positive association was observed with social attitudes towards weight and overeating in children but the estimate was attenuated with adjustment (aOR = 1.03 (95% CI 1.00, 1.07; Table 3). In these models, acculturative stress was associated with dieting ≥5 times a year (aOR = 1.61, 95% CI 1.09, 2.38), overeating (aOR = 2.37, 95% CI 1.60, 3.51), and compensatory behaviors (aOR = 2.02, 95% CI 1.14, 3.60). Acculturative stress was therefore retained in models for both adolescents and children; however, Hispanic/Latino background and AHIMSA category for acculturation were not as they did not significantly change our estimates.
Additionally, a positive association was observed between social attitudes towards weight and BMI percentile among adolescents (adjusted β = 0.63, 95% CI 0.28, 0.98; Table 5), such that greater social influences were associated with higher BMI percentiles. Results from sensitivity analysis showed that effect estimates did not change substantially when restricting ages for children and adolescent cards to ≤12 and ≥ 12 years, respectively. However, confidence intervals were wider due to the smaller sample size (see Supplemental Table).
4. Discussion
Understanding the role of body image and social ideals among Hispanic/Latino youths is necessary for improved screening of disordered eating behaviors. Consistent with our hypothesis, our results indicate that body image discrepancy is significantly associated with disordered eating behaviors among Hispanic/Latino youth, including dieting, overeating, and compensatory behaviors. Specifically, associations were stronger among individuals with a negative body image discrepancy, or a perceived actual body size larger than the perceived ideal body size, highlighting the importance of the direction of the body image discrepancy. Screening for negative body image discrepancy may improve identification of Hispanic/Latino youth with disordered eating.
Social influences on the youth’s attitudes towards weight was not associated with disordered eating behaviors as we hypothesized based on previous studies (Snapp, 2009). This suggests that Hispanic/Latino youth may have other influences outside of the media and American culture that influence their attitudes towards weight and body image. Another reason for our null findings may be that the items of the internalization subscale of the SATAQ do not appropriately capture important social influences in our sample. For example, the SATAQ does not have items related to social media influences, which may be more influential in our sample than movies and magazines.
Using a large sample of Hispanic/Latino youth from four U.S. urban areas with the highest Hispanic/Latino concentrations, we were able to observe some differences by Hispanic/Latino background, although the sample size was small for most groups. Distinct differences were observed among adolescents by Hispanic/Latino background with Cubans being the only group reporting a null mean body image discrepancy and having the lowest BMI percentile and Mexicans having the highest proportion reporting overeating. To our knowledge, this is the first study reporting differences in body image discrepancy and disordered eating by Hispanic/Latino background. Additionally, this is the first study to examine this association in a large, diverse group of Hispanic/Latinos and adjust for the effect of acculturative stress, factors that may bias other studies.
In regression models, we found that acculturative stress was a significant predictor, while Hispanic/Latino background and acculturation using the AHIMSA categories for acculturation were not. This may be due to the poor internal consistency across the groups of the AHIMSA or the fact that the Acculturative Stress Index may be more sensitive to negative pressures that come with acculturation rather than the AHIMSA, which focuses more on ethnic identity. After adjustment for acculturative stress and other covariates, all significant effect estimates remained significant, suggesting that body size perceptions may play an independent role in eating disorder etiology among Hispanic/Latino youth. However, residual confounding by acculturation level may be present and may explain null associations with social attitudes and disordered eating. We also theorize that Hispanic/Latino background was not a significant covariate in our models as it is strongly related to study site.
Significant differences by sex were also noted for body image discrepancy and social attitudes towards weight among adolescents, which were in agreement with previous studies (Ayala et al., 2007; Neumark-Sztainer, Croll, et al., 2002; Robinson et al., 1996). However, significant differences were not noted among children and sex was not a moderator in logistic regression models among both children and adolescents, but was a strong confounder. We therefore did not stratify our results by sex and instead included the variable in the regression models. It is possible that we did not observe the moderating effects by sex other studies have reported because our sample is younger compared to other studies, which focus mostly on adolescents and young adults.
We also compared these results with BMI percentiles and found that body image discrepancy was also associated with higher BMI among Hispanic/Latino youth, supporting results from previous studies (Ayala et al., 2007; Neumark-Sztainer, Croll, et al., 2002). In the disordered eating models, we did not find interaction between BMI and body image discrepancy. However, there may be other interactions we did not consider that may bias our results. Future studies should examine this further, and still consider interaction with body image discrepancy by gender and BMI in other populations. Limitations of our study include the use of cross-sectional data and the use of different scales for the child and adolescent cards assessing body size. The use of the interviewer’s judgment in assigning a child or adolescent body image card at time of interview was also a limitation; however, these measures and procedures have been previously validated (Allison, 1995; Ayala et al., 2007). Our analyses were stratified due to the different scales in each card, however, ages ranged from 8 to 16 years in both groups. While this approach was meant to account for youth entering puberty at different ages, it is possible that interviewers were more inclined to show heavier weight youths the adolescent card, although BMI percentiles were similar between both groups Because analyses considered the relative differences between perceived ideal and actual body size, effects due to the differences in scale items (i.e, 7 vs. 9 body shapes) are very minor. The majority of youth assigned the adolescent card (72.6%) were ≥12 years, and the majority of youth assigned the child card (70.2%) were ≤12 years. Sensitivity analyses restricting each card to this age cut-off did not materially alter our results. While this is the most diverse study on disordered eating conducted in Hispanic/Latino youths, the study was still underpowered to detect significant differences by Hispanic/Latino backgrounds. Nonetheless, the distribution of our variables differed by Hispanic/Latino background, which supports further refining background groups in other large national studies.
We noted a higher proportion of disordered eating behaviors among youths that were assigned a child card compared to those assigned an adolescent card, however these differences were not statistically significant. Nonetheless, these results are contradictory to the literature showing higher prevalence of disordered eating among adolescents. A possible explanation may be that youth with disordered eating behaviors may be smaller and therefore more likely to be assigned a child card.
4.1. Conclusion
In summary, this study contributes new knowledge to understand the role of body image discrepancy and disordered eating behaviors among Hispanic/Latino youth. SOL Youth, a large multi-site study of Hispanic/Latino youth with an objective of assessing risk factors for obesity, provided the ideal data for our study. Body image discrepancy was associated with disordered eating behaviors among Hispanic/Latino youth, even after adjusting for influences due to acculturation. Associations were stronger among adolescents and were present for dieting, overeating, and compensatory behaviors. A better understanding of factors contributing to body image discrepancy can help in screening for disordered eating among Hispanic/Latino youth in the United States. Differences by Hispanic/Latino background can help inform interventions targeted at certain background groups.
Supplementary Material
Acknowledgements
The authors thank the participants and staff of the HCSH/SOL and HCHS/SOL Youth Study for their important contributions. A complete list of staff and investigators is available on the study website (http://www.cscc.unc.edu/hchs/). The SOL Youth Study was supported by Grant number R01HL102130 from the National Heart, Lung, and Blood Institute. The children in SOL Youth are drawn from the study of adults: The Hispanic Children’s Community Health Study/Study of Latinos, which was supported by contracts from the National Heart, Lung, and Blood Institute (NHLBI) to the University of North Carolina (N01-HC65233), University of Miami (N01-HC65234), Albert Einstein College of Medicine (N01-HC65235), Northwestern University (N01-HC65236), and San Diego State University (N01-HC65237). The following Institutes/Centers/Offices contribute to the HCHS/SOL through a transfer of funds to NHLBI: National Center on Minority Health and Health Disparities, the National Institute of Deafness and Other Communications Disorders, the National Institute of Dental and Craniofacial Research, the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute of Neurological Disorders and Stroke, and the Office of Dietary Supplements. Additional support was provided by the Life Course Methodology Core of the New York Regional Center for Diabetes Translation Research (DK111022–8786). The study sponsors did not have any role in study design; collection, analysis, and interpretation of data; writing the report; and the decision to submit the report for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health.
Footnotes
Ethical statement
Institutional review boards (IRBs) at each study center (San Diego State University, University of Illinois Chicago, Albert Einstein College of Medicine, and University of Miami) reviewed and approved this study. All participants gave informed consent, or assent if under the age of 7 years, before taking part in this study.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.appet.2022.106079.
References
- Allison DB (1995). Handbook of assessment methods for eating behaviors and weight related problems: Measures, theory, and reasearch. Thousand Oaks, CA: Sage. [Google Scholar]
- Austin JL, & Smith JE (2008). Thin ideal internalization in Mexican girls: A test of the sociocultural model of eating disorders. International Journal of Eating Disorders, 41(5), 448–457. [DOI] [PubMed] [Google Scholar]
- Ayala GX, Mickens L, Galindo P, & Elder JP (2007). Acculturation and body image perception among Latino youth. Ethnicity and Health, 12(1), 21–41. 10.1080/13557850600824294 [DOI] [PubMed] [Google Scholar]
- Barry DT (2005). Measuring acculturation among male Arab immigrants in the United States: An exploratory study. Journal of Immigrant Health, 7(3), 179–184. [DOI] [PubMed] [Google Scholar]
- Croll J, Neumark-Sztainer D, Story M, & Ireland M (2002). Prevalence and risk and protective factors related to disordered eating behaviors among adolescents: Relationship to gender and ethnicity. Journal of Adolescent Health, 31(2), 166–175. from http://www.ncbi.nlm.nih.gov/pubmed/12127387. [DOI] [PubMed] [Google Scholar]
- Erickson SJ, & Gerstle M (2007). Investigation of ethnic differences in body image between Hispanic/biethnic-Hispanic and non-Hispanic White preadolescent girls. Body Image, 4(1), 69–78. [DOI] [PubMed] [Google Scholar]
- Fitzgibbon ML, Blackman LR, & Avellone ME (2000). The relationship between body image discrepancy and body mass index across ethnic groups. Obesity Research, 8(8), 582–589. 10.1038/oby.2000.75 [DOI] [PubMed] [Google Scholar]
- Forney KJ, & Ward RM (2013). Examining the moderating role of social norms between body dissatisfaction and disordered eating in college students. Eating Behaviors, 14(1), 73–78. [DOI] [PubMed] [Google Scholar]
- Franko DL, Coen EJ, Roehrig JP, Rodgers RF, Jenkins A, Lovering ME, et al. (2012). Considering J. Lo and ugly betty: A qualitative examination of risk factors and prevention targets for body dissatisfaction, eating disorders, and obesity in young Latina women. Body Image, 9(3), 381–387. [DOI] [PubMed] [Google Scholar]
- Gil AG, & Vega WA (1996). Two different worlds: Acculturation stress and adaptation among Cuban and Nicaraguan families. Journal of Social and Personal Relationships, 13(3), 435–456. 10.1177/0265407596133008 [DOI] [Google Scholar]
- Gowen LK, Hayward C, Killen JD, Robinson TN, & Taylor CB (1999). Acculturation and eating disorder symptoms in adolescent girls. Journal of Research on Adolescence, 9(1), 67–83. 10.1207/s15327795jra0901_4 [DOI] [Google Scholar]
- Grabe S, Ward LM, & Hyde JS (2008). The role of the media in body image concerns among women: A meta-analysis of experimental and correlational studies. Psychological Bulletin, 134(3), 460–476. 10.1037/0033-2909.134.3.460 [DOI] [PubMed] [Google Scholar]
- Hales CM, Carroll MD, Fryar CD, & Ogden CL (2017). Prevalence of obesity among adults and youth: United States, 2015-2016. NCHS data brief (Vol. 288, pp. 1–8). from http://www.ncbi.nlm.nih.gov/pubmed/29155689. [PubMed] [Google Scholar]
- Heinberg LJ, & Thompson JK (1995). Body image and televised images of thinness and attractiveness: A controlled laboratory investigation. Journal of Social and Clinical Psychology, 14(4), 325–338. 10.1521/jscp.1995.14.4.325 [DOI] [Google Scholar]
- Isasi CR, Carnethon MR, Ayala GX, Arredondo E, Bangdiwala SI, Daviglus ML, … Van Horn L (2014). The hispanic community children’s health study/study of Latino youth (SOL youth): Design, objectives, and procedures. Annals of Epidemiology, 24(1), 29–35. 10.1016/j.annepidem.2013.08.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Labre MP (2002). Adolescent boys and the muscular male body ideal. Journal of Adolescent Health, 30(4), 233–242. from http://www.ncbi.nlm.nih.gov/pubmed/11927235. [DOI] [PubMed] [Google Scholar]
- Lopez V, Corona R, & Halfond R (2013). Effects of gender, media influences, and traditional gender role orientation on disordered eating and appearance concerns among Latino adolescents. Journal of Adolescence, 36(4), 727–736. 10.1016/j.adolescence.2013.05.005 [DOI] [PubMed] [Google Scholar]
- Neumark-Sztainer D, Croll J, Story M, Hannan PJ, French SA, & Perry C (2002). Ethnic/racial differences in weight-related concerns and behaviors among adolescent girls and boys: Findings from project EAT. Journal of Psychosomatic Research, 53(5), 963–974. from http://www.ncbi.nlm.nih.gov/pubmed/12445586. [DOI] [PubMed] [Google Scholar]
- Neumark-Sztainer D, Paxton SJ, Hannan PJ, Haines J, & Story M (2006). Does body satisfaction matter? Five-year longitudinal associations between body satisfaction and health behaviors in adolescent females and males. Journal of Adolescent Health, 39(2), 244–251. 10.1016/j.jadohealth.2005.12.001 [DOI] [PubMed] [Google Scholar]
- Neumark-Sztainer D, Story M, Hannan PJ, Perry CL, & Irving LM (2002). Weight-related concerns and behaviors among overweight and nonoverweight adolescents: Implications for preventing weight-related disorders. Archives of Pediatrics and Adolescent Medicine, 156(2), 171–178. from http://www.ncbi.nlm.nih.gov/pubmed/11814380. [DOI] [PubMed] [Google Scholar]
- Neumark-Sztainer D, Story M, Resnick MD, & Blum RW (1998). Lessons learned about adolescent nutrition from the Minnesota adolescent health survey. Journal of the American Dietetic Association, 98(12), 1449–1456. 10.1016/S0002-8223(98)00329-0 [DOI] [PubMed] [Google Scholar]
- Perreira KM, Marchante AN, Schwartz SJ, Isasi CR, Carnethon MR, Corliss HL, … Delamater AM (2019). Stress and resilience: Key correlates of mental health and substance use in the hispanic community health study of Latino youth. Journal of Immigrant and Minority Health, 21(1), 4–13. 10.1007/s10903-018-0724-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pike KM, Hoek HW, & Dunne PE (2014). Cultural trends and eating disorders. Current Opinion in Psychiatry, 27(6), 436–442. [DOI] [PubMed] [Google Scholar]
- Quick VM, & Byrd-Bredbenner C (2014). Disordered eating, socio-cultural media influencers, body image, and psychological factors among a racially/ethnically diverse population of college women. Eating Behaviors, 15(1), 37–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Robinson TN, Killen JD, Litt IF, Hammer LD, Wilson DM, Haydel KF, … Taylor CB (1996). Ethnicity and body dissatisfaction: Are hispanic and asian girls at increased risk for eating disorders? Journal of Adolescent Health, 19(6), 384–393. from http://www.ncbi.nlm.nih.gov/pubmed/8969369. [DOI] [PubMed] [Google Scholar]
- Smart JF, & Smart DW (1995). Acculturative stress: The experience of the Hispanic immigrant. The Counseling Psychologist, 23(1), 25–42. [Google Scholar]
- Snapp S (2009). Internalization of the thin ideal among low-income ethnic minority adolescent girls. Body Image, 6(4), 311–314. 10.1016/j.bodyim.2009.05.005 [DOI] [PubMed] [Google Scholar]
- Swanson SA, Crow SJ, Le Grange D, Swendsen J, & Merikangas KR (2011). Prevalence and correlates of eating disorders in adolescents. Results from the national comorbidity survey replication adolescent supplement. Archives of General Psychiatry, 68(7), 714–723. 10.1001/archgenpsychiatry.2011.22 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Unger JB, Gallaher P, Shakib S, Ritt-Olson A, Palmer PH, & Johnson CA (2002). The AHIMSA acculturation scale::A new measure of acculturation for adolescents in a multicultural society. The Journal of Early Adolescence, 22(3), 225–251. 10.1177/02731602022003001 [DOI] [Google Scholar]
- Unger JB, Ritt-Olson A, Wagner K, Soto D, & Baezconde-Garbanati L (2007). A comparison of acculturation measures among Hispanic/Latino adolescents. Journal of Youth and Adolescence, 36(4), 555–565. [Google Scholar]
- Vander Wal JS, Gibbons JL, & Grazioso M (2008). The sociocultural model of eating disorder development: Application to a Guatemalan sample. Eating Behaviors, 9(3), 277–284. 10.1016/j.eatbeh.2007.10.002 [DOI] [PubMed] [Google Scholar]
- VanderWeele TJ, & Robins JM (2007). Directed acyclic graphs, sufficient causes, and the properties of conditioning on a common effect. American Journal of Epidemiology, 166(9), 1096–1104. [DOI] [PubMed] [Google Scholar]
- Vartanian LR (2012). Self-discrepancy theory and body image. Encyclopedia of body image and human appearance, 2(1), 711–717. [Google Scholar]
- Warren CS, Gleaves DH, & Rakhkovskaya LM (2013). Score reliability and factor similarity of the Sociocultural Attitudes towards Appearance Questionnaire-3 (SATAQ-3) among four ethnic groups. Journal of eating disorders, 1(1), 14. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.