The relation between weight-based teasing and psychological adjustment in adolescents

Abstract

OBJECTIVE:

To determine the prevalence of weight-based teasing, and evaluate its association with depression, anxiety and unhealthy eating behaviour in a large sample of adolescents in the Ottawa (Ontario) area.

METHODS:

A total of 1491 adolescents from public and private middle schools and high schools in rural and urban areas of Ottawa responded confidentially to surveys.

RESULTS:

More girls than boys reported that they experienced weight-based teasing (33% versus 18%). The prevalence of weight-based teasing by peers was significantly higher among overweight and obese youth than among normal weight youth (45% versus 22%). Teasing about body weight was consistently associated with anxiety, psychological distress and disordered eating, and these associations held for both boys and girls, and were independent of weight status.

CONCLUSIONS:

Weight-based teasing is a common experience among Ottawa-area adolescents, especially among overweight girls, and was found to be associated with psychological morbidity. Effective interventions are needed to help victims cope with and prevent further weight-based teasing and its harmful psychological sequelae.

OBJECTIF :

Déterminer la prévalence des moqueries liées au poids et en évaluer l'association avec la dépression, l'anxiété et les comportements alimentaires malsains dans un vaste échantillon d'adolescents de la région d'Ottawa, en Ontario.

MÉTHODOLOGIE :

Au total, 1 491 adolescents d’écoles secondaires publiques et privées des régions rurales et urbaines d'Ottawa ont répondu confidentiellement au sondage.

RÉSULTATS :

Plus de filles que de garçons ont déclaré subir des moqueries liées à leur poids (33 % par rapport à 18 %). La prévalence des moqueries liées au poids par les camarades était considérablement plus élevée chez les jeunes obèses ou faisant de l'embonpoint que chez ceux qui avaient un poids normal (45 % par rapport à 22 %). Les moqueries liées au poids corporel s'associaient de façon constante à l'anxiété, à la détresse psychologique et à une alimentation désordonnée, et ces associations s'observaient tant chez les garçons que chez les filles, quel que soit leur poids.

CONCLUSIONS :

Les moqueries liées au poids sont courantes chez les adolescents de la région d'Ottawa, notamment chez les filles faisant de l'embonpoint, et il a été établi qu'elles s'associent à une morbidité psychologique. Des interventions efficaces s'imposent pour aider les victimes à affronter et à prévenir la poursuite des moqueries liées au poids et ses séquelles psychologiques néfastes.

Obesity in childhood and adolescence is reaching epidemic proportions, and represents an important public health concern given its association with adverse medical and psychosocial outcomes (1). In general, the psychosocial outcomes have received less attention than the adverse medical outcomes (2). One aspect of psychosocial problems associated with obesity in children and youth that has been receiving increased scientific attention is weight bias and discrimination (3). One form of weight bias is weight-based teasing by peers and/or family members, but this area has not received a lot of scientific inquiry. Teasing has been defined as a personal communication from an agent to a target that combines elements of humour, aggression or ambiguity (4).

As expected, weight-based teasing occurs more often in overweight children and youth than in their normal weight peers (5–9). Much of the literature on weight-based teasing has shown untoward effects on body dissatisfaction, self-esteem and disordered eating, including bulimic eating behaviour (6,10,11). However, many studies relied on clinical samples in adults (12–14) and virtually all were conducted in the United States; thus, the results may not be generalizable to adolescents in the Ottawa (Ontario) area. In a large study (15) of Canadian children, a strong association was found between aggression (in the form of bullying) and overweight and obese school children. However, teasing was included into broader categories of peer victimization and bullying; thus, the prevalence and possible untoward psychological sequelae of weight teasing per se remains unknown in Ottawa-area children.

The social and psychological ramifications of weight teasing may hinder the social development of overweight and obese youth because adolescents are extremely reliant on peers for social support, identity formation and self-esteem. A landmark study by Eisenberg et al (5) found that weight-based teasing was consistently associated with body dissatisfaction, low self-esteem, depressive symptoms and even suicidal ideation. These associations held for boys and girls after controlling for ethnicity and even body weight. The finding of suicidal ideation highlights the potentially traumatic psychological effects of peer victimization that often accompany weight-based teasing. A more recent study (10) found that weight-based teasing by family members was associated with unhealthy weight control behaviours and poorer psychological functioning.

While it may be intuitive that weight-based teasing can elicit, perpetuate or magnify anxiety among adolescents, very little research has examined these associations. Young-Hyman et al (7) found that trait anxiety was associated with peer teasing. Trait anxiety is a more enduring form of anxiety that is characteristic of personality style, and is differentiated from state anxiety, which is more situation specific (ie, anxiety felt at a moment in time). Similarly, a relationship between weight-based teasing and emotional eating has been anecdotally reported and could very well involve a vicious cycle in which teasing leads to emotional eating, and emotional eating leads to weight gain and further weight-based teasing, but this has not been scientifically investigated.

The present study builds on past research by examining the prevalence of weight-based teasing in an Ottawa-area sample of adolescents, and serves to elucidate the relationship between weight-based teasing from peers and family members and a broader spectrum of outcome variables, such as depression, anxiety and eating behaviour.

METHODS

Participants

Questionnaires were administered to 1590 youth (683 boys and 907 girls), and 1491 were returned fully completed. The sample for the present study was drawn from 24 middle schools and high schools in the Ottawa and surrounding regions. Students who participated in the study were English-speaking boys and girls from grade 7 to grade 12, and the data were collected between November 2004 and May 2006. The students received an information presentation in their class outlining the purpose of the study and the consent process. Students were then given a consent form to take home to parents to sign. If students returned a signed consent form from their parents, they were welcome to participate. Students also signed an assent form in class before completing the questionnaire.

The research team established firm liaisons with three school boards (Ottawa-Carleton District and Catholic School Boards, and Upper Canada District School Board) and several private schools in the Ottawa area. To achieve an accurate representative sample of students from the Ottawa region, schools were geographically dispersed, with some participating schools being drawn from urban, suburban and inner-city locations. Approval to conduct the present study was obtained from the research ethics board at the Children's Hospital of Eastern Ontario (Ottawa), the Ottawa-Carleton District and Catholic School Boards, and the Upper Canada District School Board.

Procedures

Students completed surveys in classrooms in one session of 75 min. Research personnel were on hand to answer any questions or deal with any critical incidents such as suicidality. All completed questionnaires were sealed in an envelope by students and deposited in a box at the front of the classrooms to ensure confidentiality of responses.

Measures

Demographic data were collected from each participant, including details on grade, date of birth, age and sex.

Weight measurements were collected using a standardized digital scale (Health-O-Meter, model 830 KL; Sunbeam Products Inc, USA) and recorded in kilograms to the nearest 0.1 kg. Height measurements were obtained using a stadiometer (model 217; Seca Corp, USA) and were recorded in centimetres to the nearest 0.1 cm. Students were asked to remove their shoes before measurements were taken. Weight was measured privately and individually in a separate room. Body mass index (BMI) was calculated by dividing the participants’ weight in kilograms by their height in metres squared (kg/m²). BMI percentile was calculated based on growth charts from the Centers for Disease Control and Prevention (USA) (16) and used to classify participants into normal weight, overweight or obese categories.

The Dutch Eating Behaviour Questionnaire (17) is a 33-item self-report questionnaire that contains 5-point Likert-type scales ranging from 1 (never) to 5 (very often). The survey has three subscales: restrained eating (10 items), emotional eating (13 items) and external eating (10 items). Negative items are reversed so that a high score always reflects a positive value judgment of the body. The reported Cronbach's alpha coefficients for the varying subscales ranged from 0.79 to 0.95. It has been validated with adult and youth populations, and when measured with young girls, the computation of internal consistency showed an alpha coefficient of 0.83 for Dutch Eating Behaviour Questionnaire total, and subscale alpha values ranging from 0.77 to 0.86 (18,19).

The Child Depression Inventory (20) is a 27-item questionnaire that asks six- to 17-year-old respondents to endorse statements about themselves reflecting cognitive, behavioural and somatic symptoms of depression. The items are rated on a 3-point scale indicating symptom severity (ie, 0 = no presence of symptoms, and 2 = highest severity possible). Total scores on the Child Depression Inventory range from 0 to 52, with higher scores indicative of greater reports of depressive symptomatology. The measure has been shown to have acceptable internal consistency (alphas ranging from 0.71 to 0.89) and good discriminant validity when classifying children with no significant psychopathology versus those who are depressed (20).

Weight-based teasing was assessed by the McKnight Risk Factor Survey III (21). The peer teasing subscale consists of eight items and the parent teasing subscale consists of three items. Each item uses a 5-point scale anchored by 1 (never) to 5 (always). Parent and peer teasing subscales were used in the correlational analyses. Prevalence of teasing by peers and adults was operationally defined by any response of ‘a little’ teasing or more. The test-retest reliability, internal consistency and convergent validity of the McKnight survey are well established.

The Multidimensional Anxiety Scale for Children (MASC-10) (22) is a brief 10-item scale adapted from the original 39-item MASC that assesses overall levels of anxiety symptomatology in eight- to 19-year-old youth. Respondents are asked to rate each symptom using a 4-point Likert scale ranging from 0 (never true about me) to 3 (often true about me). Test-retest reliability for the adolescent sample (13- to 18-year-old adolescents) falls in the satisfactory to excellent range, with interclass correlation coefficients ranging from 0.75 to 0.92 (22). The total interclass correlation coefficient for this sample was 0.88. As for convergent validity, the MASC total score showed significant correlations with the Revised Children's Manifest Anxiety Scale with r=0.633 (22,23).

Analysis

χ² analyses were used to compare the prevalence of weight-based teasing in the overall sample and by weight status in boys versus girls. Normal weight status was defined as those with a BMI ranging from higher than the fifth to lower than the 85th percentile. Overweight was defined as having a BMI from the 85th to lower than the 95th BMI percentile, while obesity was defined as having a BMI at the 95th percentile or higher for age and sex. Pearson correlations, partialling out BMI, between weight-based teasing by parents and peers and psychosocial variables of interest were conducted by sex and weight status. Although not all of the variables were normally distributed, the pattern of correlations using nonparametric Spearman correlations was the same; thus, values for Pearson correlations are presented.

RESULTS

Of the 1590 questionnaires administered, 1491 were returned fully completed, making the response rate a respectable 94%. Table 1 shows the demographic and anthropometric characteristics of the sample. The prevalence of overweight and obesity (ie, 24%) found in the present study closely matches the prevalence found in nationally representative surveys (1).

The prevalence of experienced weight-based teasing by peers in the overall sample was 29%. More girls than boys reported being teased about their weight (33% versus 18%; P<0.001). Regarding weight status, the prevalence of weight-based teasing by peers was significantly higher among overweight and obese youth than among normal weight youth (45% versus 22%; P<0.001). Overweight and obese girls experienced more peer weight teasing than overweight and obese boys (52% versus 30%; P<0.001).

A total of 21% of adolescents reported being teased by parents about their weight, with girls reporting higher prevalences than boys (25% versus 15%; P<0.01). The prevalence of weight teasing by adults was significantly higher in overweight and obese youth than in normal weight youth (25% versus 18%; P<0.01). Overweight girls reported higher prevalences of parent teasing than overweight and obese boys (30% versus 22%; P<0.001).

Table 2 shows the relationship between weight-based teasing by peers and parents, and various aspects of psychosocial functioning, with BMI statistically controlled. Weight teasing by both parents and peers was significantly associated with psychosocial maladjustment and disordered eating behaviour.

Table 3 shows the relationship between parent and peer teasing and psychological adjustment in normal weight, overweight and obese adolescents. Results indicate that weight status did not have a large impact on the relationship between parent teasing and psychological adjustment, because correlations with psychological factors were similar among normal weight, overweight and obese youth. However, the association between weight teasing by peers and psychological maladjustment was consistently higher among obese youth than normal weight youth.

DISCUSSION

To our knowledge, the present study is the first to examine the prevalence and correlates of weight-based teasing in a large school-based sample of adolescents in Canada. Similar to the American-based studies, we found that the experience of weight-based teasing is prevalent, especially among those who are overweight or obese. In both normal weight and overweight classifications, girls reported higher prevalences of weight-based teasing than boys – a sex difference that has been reported in American studies (7,8).

The clinical significance of our findings is derived by the strong correlation between teasing and psychological variables of interest. In both boys and girls, being teased was associated with subscales on unhealthy eating behaviours, such as dietary restraint, emotional eating and external eating, consistent with American-based studies (6,8,11). The finding that emotional eating was a correlate of weight teasing is interesting. It suggests that youth may be using food to cope with the teasing, which increases the chance of weight gain, and increased weight gain may increase the risk of weight-based teasing, creating a vicious cycle. This hypothesis needs to be tested in subsequent prospective research.

We found a consistently significant association between weight teasing by parents and peers and various aspects of depression, including negative mood, interpersonal problems, anhedonia, negative self-esteem and total score of depression. These findings are consistent with a large school-based study in the Unites States (5), and are particularly alarming given the potential debilitating effects of depression.

To the best of our knowledge, we are among the first to establish a link between weight-based teasing and anxiety among Canadian boys and girls. Teasing is a form of victimization so it is easy to understand how this may lead to anxiety, but it is possible that youth who exhibit signs of anxiety may be targets for teasing; again, creating a vicious cycle.

It is interesting to note that even though overweight and obese youth were teased about their weight more than their nonoverweight peers, the association with psychological distress was similar across weight status. While teasing even in the normal weight youth was associated with negative psychological outcomes, this suggests that it is the experience of being teased about weight, rather than actual body shape and weight, that is relevant to unhealthy eating behaviour, depressed mood and anxiety (5).

Of notable importance, very little research has comparatively examined the psychological effects of parent versus peer teasing in children and adolescents. Eisenberg et al (5) found that the combination of peer and parent teasing was associated with more psychological distress than either form of teasing on its own. In the present study, the relationship between teasing, weight status and psychological adjustment was influenced by the perpetrator of the teasing: parents or peers. Our sample consisted of adolescents, and the teenage years are often marked by a shift in influence from parents to peers. Although we did not specifically test this, the correlations between weight teasing and psychological morbidity appear to be higher for peers than for parents, but only in the overweight and obese youth and not for the normal weight youth. It is well known that obese youth have weaker social support systems and tend to be socially marginalized more than nonoverweight youth (3); thus, the experience of being teased by peers may have a more psychologically damaging effect. Given that parents have a stronger influence during the developmental preadolescent years, weight-based teasing by parents may be more harmful than teasing by peers before adolescence. Future research is needed to elucidate which form of teasing has more untoward effects, and how stage of development, sex and personality variables moderate the psychological effects of teasing. This information would aid the development of weight-based teasing intervention programs.

This research has strengths and limitations that need to be considered. The strengths of the study included using inventories that have well-established psychometric properties in adolescents. The heights and weights were objectively measured to calculate weight status rather than relying on self-report, which is common in most large surveys. Finally, we believe we are the first in Canada to report on weight-based teasing and its association to psychosocial functioning in adolescents.

There are also limitations to our study. As previously mentioned, the data are cross-sectional; thus we cannot infer that teasing causes psychological distress in adolescents. It may well be that those who are distressed psychologically may be more sensitive and vulnerable to peer teasing than those who are not distressed. Second, all data were self-reported (except height and weight), so it is possible that participants’ perceptions of teasing experiences from peers and parents differed from actual teasing experiences, which is impossible to evaluate outside the laboratory in real-world settings. As noted in previous research (5), our measure of teasing by peers and parents was broad and subscales were composed of multiple items, thus precluding the influence of specific forms of teasing. Future research may need to examine which teasing items have the strongest association with emotional well-being, and whether peer teasing is more harmful than parent teasing, because this may influence the development of intervention programs. Moreover, broader outcomes may be examined in relation to teasing, such as substance abuse. Finally, no information was gathered for participants who did not complete the survey; thus, the degree of bias introduced in this study is unknown. However, it would be reasonable to conclude that bias would be minimal given the high (94%) response rate. Given the high response rate and broad sampling of schools that included private and public schools in urban and rural settings, we believe the sample is representative of adolescents living in the metropolitan Ottawa region. However, given that Ottawa is more affluent than many other cities in Ontario and Canada, and that only 10.8% of youth come from low-income families (24), our results may not be generalizable to all adolescents in Canada.

CONCLUSION

Our data indicate that overweight and obese adolescents in the metropolitan Ottawa region are at an increased risk of being teased about their weight, and this teasing is associated with unhealthy eating behaviours and a poor psychosocial profile characterized by elevated anxiety and depression scores. Longitudinal research is needed to establish directionality of the relationship between weight-based teasing and psychosocial difficulties. Nevertheless, the current findings have important implications for the development of screening and mental health promotion programs for youth at risk of weight-based teasing and associated unhealthy eating behaviour, anxiety and depression. For example, educators, guidance counsellors and school psychologists may want to be more vigilant of their overweight and obese students with regard to their psychological state of mind and whether they are being teased about their weight. Moreover, health promotion programs designed to educate students about healthy eating and active living would be beneficial to reduce obesity and encourage healthy weights. In addition, informing students that there are strong genetic and biological factors that influence body weight and shape (25) may help reduce some shame or guilt that obese children often feel, and this may also reduce stigma, bias and weight-based teasing that obese children often experience (3,26). Our findings suggest that overweight and obese youth may be particularly vulnerable to weight-based teasing and related mental health issues. Implementing a zero tolerance policy for weight teasing may help reduce the untoward psychological sequelae. Although the development and evaluation of programs targeting the reduction of weight-based teasing are still in its infancy (9,27), the initial school-based interventions provide empirical data that are very encouraging (27).

Data for this study were collected as part of the Research on Eating and Adolescent Lifestyles (REAL) study, which was supported by funding from the University of Ottawa Medical Research Fund, and the Ontario Provincial Centre of Excellence for Child and Youth Mental Health. The authors also thank Dr Robert Flewelling, Andrea Byrne and Andrew Lumb for their assistance with data collection and data entry.

REFERENCES