Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
Verbal harassment, such as bullying and hate speech, has received considerable attention recently, but less is known about weight-based teasing and its potential harmful effects on young people's psychosocial well-being.
To determine the associations of weight-based teasing and body satisfaction, self-esteem, depressive symptoms, and suicidal ideation and suicide attempts using a large sample of adolescents.
Secondary analysis of survey and anthropometric data.
Ethnically and socioeconomically diverse communities in the urban and suburban school districts of the Minneapolis/St Paul metropolitan area.
A school-based sample of 4746 adolescents in grades 7 to 12 at 31 public middle schools and high schools.
Main Outcome Measures
Weight-based teasing from peers or family members, body satisfaction, self-esteem, depressive symptoms, and suicidal ideation and suicide attempts.
Of the eligible students, 81.5% participated; 30.0% of adolescent girls and 24.7% of adolescent boys were teased by peers, and 28.7% of adolescent girls and 16.1% of adolescent boys were teased by family members. Approximately 14.6% of adolescent girls and 9.6% of adolescent boys reported teasing from both of these sources. Teasing about body weight was consistently associated with low body satisfaction, low self-esteem, high depressive symptoms, and thinking about and attempting suicide, even after controlling for actual body weight. These associations held for adolescent boys and girls, across racial, ethnic, and weight groups. Furthermore, teasing from 2 sources was associated with a higher prevalence of emotional health problems than either teasing from a single source or no teasing.
Physicians and other health care providers should recognize the importance of weight-based teasing for young patients. Policy, programs, and education should focus on increasing awareness of what constitutes weight-based teasing, its potentially harmful effects on adolescents' emotional well-being, and reduction of this behavior.
HARMFUL INTERPERSONAL interactions, such as bullying and hate speech, have come under scientific and public scrutiny in recent years,1- 5 and some steps have been taken to reduce their incidence and minimize their effects. Less public attention, however, has focused specifically on weight-based teasing as a potentially harmful interaction. Given the importance placed on body shape and size in US culture, weight-based teasing may pose a serious threat to the health and well-being of young people.
Weight-based teasing by peers and family members is widespread, especially for overweight individuals, and exists across racial and ethnic groups.6 Previous research7- 16 has identified teasing as one risk factor for body dissatisfaction and disordered eating; however, most studies focusing specifically on weight- or appearance-based teasing have typically used clinical10,15 or non-US14 samples, which may not be generalizable to the teasing experiences of US young people. Other studies7,8,11 used samples too small to allow for comparisons between racial and ethnic groups or across weight categories. Previous studies also do not address suicide in relation to weight-based teasing.
Depressive symptoms are widespread, affecting 10% to 15% of the child and adolescent population,17 and are a known precursor to suicidal ideation and attempts. Suicide is the third leading cause of death among adolescents,18 accounting for 13.3 deaths per 100 000 among 15- to 19-year-old adolescent boys and 2.8 deaths per 100 000 among same-age girls.19 Identifying teasing as a modifiable social factor relevant to emotional well-being and suicide may suggest strategies for prevention.
This study, therefore, uses a large school-based sample of adolescents to determine the associations of weight-based teasing and body satisfaction, self-esteem, depressive symptoms, and suicidal ideation and suicide attempts. The primary hypotheses are that young people who are teased about weight will report lower body satisfaction, lower self-esteem, and more depressive symptoms and be at an increased risk for suicidal behaviors compared with those not teased.
Data for this study come from Project EAT (Eating Among Teens), a comprehensive study of eating patterns and weight concerns among adolescents. Thirty-one public middle schools and high schools in ethnically and socioeconomically diverse communities in the urban and suburban school districts of the Minneapolis/St Paul metropolitan area participated in the study (of 55 recruited to participate). A 221-item survey assessing a range of socioeconomic, personal, and behavioral factors relevant to nutritional health was completed by students during regular class periods, and height and weight were measured by trained research staff in a private area of the school. Study procedures were approved by the University of Minnesota Human Subjects' Committee and by the research boards of the participating school districts. The response rate for student participation was 81.5%, and response rates varied across schools (28.2%-94.9%), based largely on the school's use of active vs passive parental consent. The main reasons for nonparticipation were absenteeism and failure to return consent forms. Further details of the study design and data collection methods are given elsewhere.16,20
The survey was designed by Project EAT investigators (including D.N.-S. and M.S.). Most items were drawn from existing adolescent surveys; some were developed to meet the specific needs of the study. The survey went through multiple revisions based on input from experts in the field, an adolescent advisory board, and 68 students in the 7th and 10th grades (at different schools in the same school districts as those who took the final survey) who pretested the survey. Follow-up pilot testing was conducted with 161 students during a 2-week interval to assess the psychometric properties of questions, problems with missing data, potential for bias, and the consistency of responses. After pilot testing and analysis were completed, wording on some items was changed to more accurately assess constructs, improve readability, or delete unnecessary items.
The primary independent variable tested in these analyses was teasing by peers or family members. Two separate survey items read, "Have you ever been teased or made fun of by other kids/family members because of your weight?" Possible responses were yes or no.21
Five measures of emotional well-being included the following: (1) Body satisfaction,22 assessed with 5-point Likert scales of satisfaction with 10 body parts; the composite score was the average of scores for each body part, and had a Cronbach α of .93 for adolescent boys and .92 for adolescent girls. (2) Self-esteem, measured with 6 items from the Rosenberg Self-esteem Scale23; this scale assessed level of agreement with statements such as "I certainly feel useless at times," and "on the whole, I am satisfied with myself," and had a reliability estimate of .78 for adolescent boys and girls. (3) Depressive mood,24 which assessed the extent (not at all, somewhat, or very much) to which respondents had been bothered or troubled by 6 depressive symptoms, such as feeling hopeless about the future and feeling unhappy, sad, or depressed; this scale had reliability estimates for adolescent boys of .76 and for adolescent girls of .75. (4) "Have you ever thought about killing yourself" was used to assess suicidal ideation. (5) "Have you ever tried to kill yourself" assessed a past suicide attempt. Responses to both included "yes, during the last year," "yes, more than a year ago," and "no." Both yes responses were grouped together for these analyses. The 3 continuous variables were dichotomized to compare those in the lowest quartile of body satisfaction and self-esteem and in the highest quartile of depressive symptoms with all other respondents.
Several additional variables were included in the analysis. Race or ethnicity was assessed with one item: "Do you think of yourself as (1) white, (2) black or African American, (3) Hispanic or Latino, (4) Asian American, (5) Hawaiian or Pacific Islander, or (6) American Indian or Native American?" Respondents were grouped as white or nonwhite for multivariate analysis. Five levels of socioeconomic status (SES) (low to high) were based on the highest educational level completed by either parent for most respondents. When this information was missing (n = 1058), eligibility for public assistance, eligibility for free or reduced-cost school meals, and parental employment status were used to infer SES.25,26 School level was defined as middle school (grades 7-8) vs high school (grades 9-12). The body mass index was determined by anthropometric measures of height and weight, using standardized equipment and procedures. Height was measured without shoes, and weight was measured in street clothes without heavy outerwear. The body mass index was calculated as weight in kilograms divided by the square of height in meters. Sex- and age-specific cut points for underweight, normal weight, overweight, and obesity were based on data from the Centers for Disease Control and Prevention.27
Multivariate logistic regression analyses were calculated with a generalized estimating equation to control for clustering of respondents by school. Main effects models included the 5 emotional health variables, using peer and family teasing as primary independent variables and race or ethnicity, SES, school level, and underweight and overweight (including obese) status as covariates. The interaction terms of teasing with race and body mass index (separately) were added to main effects models, based on previous findings16,20,28- 30 that weight concerns and teasing may be different across these groups. All analyses were conducted separately for male and female respondents. Analyses were conducted using SAS statistical software, version 8.12 (SAS Institute Inc, Cary, NC).
The sample was composed of 4746 students, including 2357 adolescent girls (49.7%) and 2377 adolescent boys (50.1%), plus 12 students who did not indicate their sex (0.2%) (Table 1). Approximately two thirds of the respondents were high school students, approximately half were white, and approximately half were categorized as middle class or upper middle class for SES.
Emotional well-being variables are also shown in Table 1. Adolescent girls were more likely than adolescent boys to report each of the 5 emotional health concerns.
Weight-based teasing was common in this sample for adolescent girls and boys: 30.0% of adolescent girls and 24.7% of adolescent boys reported being teased about weight by their peers, and 28.7% of adolescent girls and 16.1% of adolescent boys reported being teased about weight by someone in their families. Approximately 14.6% of adolescent girls and 9.6% of adolescent boys indicated that they had been teased about their weight by peers and family members.
Table 2 shows the proportion of respondents reporting each of the 5 emotional health concerns according to whether they were not teased at all, teased by peers only, teased by family only, or teased by peers and family. For all 5 measures for adolescent girls and boys, students who reported teasing by either peers or family members were more likely to have each emotional health problem than those not teased at all, as predicted. The proportion reporting each emotional health concern was even higher among those teased by peers and family than by either single source in almost all cases (except adolescent boys' suicide attempts). More than half of adolescent girls teased by both sources report thinking about suicide, compared with 24.7% of those who are not teased; and almost one quarter of those teased by both sources reported attempting suicide, compared with 8.5% of those not teased. Similarly, adolescent boys teased by family (with or without peer teasing) were approximately 3 times more likely to have attempted suicide than those not teased at all.
Table 3 shows odds ratios derived from multivariate logistic regression analyses, adjusting for overweight and underweight status, school level, race or ethnicity, and SES. Results indicate that respondents who were teased had odds of having each emotional health problem that were 1.39 to 2.35 times the odds for those not teased. These relationships were in the predicted direction and statistically significant in all instances except peer teasing and suicide attempts among adolescent boys. The odds of thinking about and attempting suicide are approximately twice as great among those teased by family members as the odds for those not teased by family, after controlling for all other variables in the model (odds ratio, 1.93-2.35).
Other associations between the covariates and emotional well-being are also apparent in Table 3. Interestingly, weight status was not associated with low self-esteem, suicidal ideation or attempts, or adolescent girls' depressive symptoms after controlling for teasing.
Interactions of weight-based teasing with white or nonwhite race and teasing with body mass index showed no clear pattern and were not consistently significant in the final models. To determine if the combination of all nonwhite groups masked important distinctions between some populations, further analysis was conducted to test the interactions of teasing with 5 separate racial or ethnic groups (white, black, Hispanic, Asian, and other). Findings continued to show no significant interactions by race for self-esteem, depressive symptoms, or suicidal ideation or attempts, and only minimal differences for body satisfaction among adolescent girls. Thus, the association of teasing with the emotional well-being of adolescents seems to be similar across racial or ethnic groups and across weight categories.
Teasing about body weight was consistently associated with low body satisfaction, low self-esteem, high depressive symptoms, and thinking about and attempting suicide. These associations held for adolescent boys and girls across racial and ethnic groups and weight groups. Teasing by both peers and family members was associated with very high rates of emotional health problems. Of particular concern are the alarming rates of suicidal ideation and attempts associated with weight-based teasing, which are 2 to 3 times as high among those who were teased compared with those not teased.
Although previous research16 has demonstrated that overweight and underweight youth endure more teasing about weight than youth of average weight, an important aspect of these findings is that the poor emotional health associated with teasing was similar across the weight categories (as discussed previously11). The finding that weight was not significantly associated with most outcomes after teasing was entered in multivariate models indicates that the experience of being teased about weight, rather than actual body shape, seems to be the relevant factor for low self-esteem, depressive symptoms, and suicidal ideation and attempts (weight status and teasing were independently associated with body satisfaction in multivariate analyses). Pervasive media images of thin women and muscular men have contributed to poor body image and unhealthy weight control practices.31- 34 This influence may further lead to unhealthy expectations of body shape among adolescents, in themselves and in others, which may in turn account for some teasing of healthy weight young people who do not conform to the cultural ideals. The role of exposure to idealized body images in the media in teasing, being teased, and possible psychosocial outcomes has yet to be explored in the scientific literature.
A somewhat surprising finding in this study is that teasing functions similarly across white and nonwhite groups; previous literature29,35,36 has demonstrated that black persons, in particular, tend to have a greater acceptance of larger body types. This again points to the importance of the teasing experience, which may override an otherwise healthy body image and sense of well-being, regardless of race.
This research must be viewed in light of certain limitations. First, these data were collected cross-sectionally; thus, we cannot infer that being teased about weight causes poor emotional health outcomes. It is possible that the relationship is reversed, such that young people with poor emotional health may be more sensitive to teasing than those who are emotionally healthy. For example, they could perceive benign communications as teasing or teasing experiences could be more painful and memorable and, therefore, reported more frequently among adolescents with compromised emotional health.
Second, all data were self-reported; thus, measures of teasing assessed respondents' perceptions of ever being teased about their weight by peers or family. Although the extent of actual teasing is impossible to gauge from these data, the perception of being teased may in fact be the more relevant influence on emotional well-being, regardless of actual teasing experiences. A young woman who believes she was called by a derogatory name, for example, could experience negative effects to her self-esteem even if she misheard the speaker or the slur was intended for another.
In addition, the measures of teasing were broad (peers and family), and we were, therefore, unable to distinguish specific sources of teasing (eg, same- or opposite-sex peers or acquaintances, parents, or siblings). Future research may assess teasing experiences more precisely, and further elucidate their implications for adolescents' emotional well-being. Finally, no information was available for the students who did not complete the survey because of absenteeism or lack of consent. The extent to which this may introduce bias into the sample is unknown. The overall response rate for the study, however, was high (81.5%).
Despite the limitations outlined, this study has several strengths that improve on previous research regarding teasing. First, this research uses a large school-based sample of ethnically and socioeconomically diverse students. The sample size and adequate minority representation made it possible to conduct valid comparisons between racial groups and examine possible interactions of race or ethnicity with teasing. Second, this study uses strong measures of emotional well-being; scales for self-esteem and depressive symptoms have been validated previously in studies23,24 of adolescents, and the single-item measures of suicidal ideation and attempts have been used in large studies37 with this population. Finally, to our knowledge, the measures of teasing used in Project EAT have not been reported on previously in a large study of adolescent health.
Historically, little attention has been paid to verbal harassment among youth, and even less to weight-based teasing specifically. This study suggests that weight-based teasing may be damaging to the emotional well-being of adolescents. These findings and previous work show potential implications for weight-specific harmful behaviors (such as disordered eating16) and for more global psychosocial health. In addition to depression and suicide behaviors, teasing may also have implications for other high-risk behaviors (such as substance use), which should be investigated further.
Weight-based teasing may occur without family members or peers realizing it. Although certain remarks may not be intended as derogatory by the speaker, they may in fact be damaging to the listener. Therefore, a first critical step in the prevention of this behavior would include identifying the type of language and comments that constitute weight-based teasing. A second step would entail increasing awareness about the potential severe consequences of this teasing. Parents, teachers, school personnel, other youth workers, and young people may not realize the strength of such comments, and could benefit from education on this topic.
Physicians and other health care providers and health educators should recognize the potential importance of weight-based teasing for their young patients and actively counter unrealistic norms regarding body weight and shape. They should inquire about the teasing experiences of children and adolescents, including if they have been teased, how much it bothered them, and how they dealt with it. Physicians and other health care providers may want to discuss the prevalence of teasing and its unacceptability, and advocate a more reasonable and healthy weight standard for their young patients and their families.
Because the complete elimination of a widely prevalent behavior such as weight-based teasing is unlikely, additional work needs to be done with youth who may experience teasing to provide them with skills to deal with this harmful speech. Individual clinic sessions, support groups, and community-based education should emphasize skills for responding to teasers and coping with these negative messages in a healthful way.
Corresponding author and reprints: Marla E. Eisenberg, ScD, MPH, Division of General Pediatrics and Adolescent Health, School of Public Health, University of Minnesota, 200 Oak St SE, Suite 260, Minneapolis, MN 55455 (e-mail: firstname.lastname@example.org).
Accepted for publication November 14, 2002.
This study was supported by grants MCJ-270834 (Dr Neumark-Sztainer) and 5-T71-MC00006-24-Leadership Education in Adolescent Health from the Maternal and Child Health Bureau, Health Resources and Service Administration, US Department of Health and Human Services, Rockville, Md; and grant U48/CCU513331 from the Centers for Disease Control and Prevention, Atlanta, Ga.
Little public or scientific attention has focused on weight-based teasing as a potentially harmful influence. Given the importance placed on body shape and size in US culture, weight-based teasing may pose a serious threat to the health and well-being of young people.
This study demonstrates that teasing about body weight was consistently associated with low body satisfaction, low self-esteem, high depressive symptoms, and thinking about and attempting suicide among adolescents. Of particular concern are the high rates of suicidal ideation and attempts associated with weight-based teasing, which are 2 to 3 times as high among those who were teased compared with those not teased. Physicians and other health care providers and health educators should recognize the potential importance of weight-based teasing for their young patients and actively counter unrealistic norms regarding body weight and shape.
Eisenberg ME, Neumark-Sztainer D, Story M. Associations of Weight-Based Teasing and Emotional Well-being Among Adolescents. Arch Pediatr Adolesc Med. 2003;157(8):733-738. doi:10.1001/archpedi.157.8.733