 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Metabolism and Obesity Services, Royal Prince Alfred Hospital, Sydney, Australia
b Human Nutrition Unit
c Discipline of Medicine, University of Sydney, Sydney, Australia
d Academic Unit of Psychiatry and Behavioural Sciences, School of Medicine, University of Leeds, Leeds, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PARTICIPANTS AND METHODS. A total of 2813 children (mean age: 11.3 years) took part in the study. They were recruited from 55 schools and were all in the last 2 years of primary school. Participants completed the Self-perception Profile for Children, a measure of body shape perception, and their height and weight were measured.
RESULTS. Obese children had significantly lower perceived athletic competence, physical appearance, and global self-worth than their normal weight peers. Obese girls scored lower in these domains than obese boys and also had reduced perceived social acceptance. Obese children were 2–4 times more likely than their normal weight peers to have low domain competence. In terms of prevalence, 1 of 3 obese boys and 2 of 3 obese girls had low appearance competence, and 10% and 20%, respectively, had low global self-worth. Body dissatisfaction mediated most of the association between BMI and low competence in boys but not in girls.
CONCLUSIONS. Obesity impacts the self-perception of children entering adolescence, especially in girls, but in selected areas of competence. Obese children are at particular risk of low perceived competence in sports, physical appearance, and peer engagement. Not all obese children are affected, although the reasons for their resilience are unclear. Quantifying risk of psychological distress alongside biomedical risk should help in arguing for more resources in child obesity treatment.
Key Words: child obesity • self-esteem • athletic competence • physical appearance • body dissatisfaction
Abbreviations: SES—socioeconomic status
Because childhood obesity has become a focus for preventative and management action, children's psychological state has come under increasing attention. Concerns about body dissatisfaction and low self-esteem in obese children and adolescents, especially girls, center on their distance from societal body shape ideals and experience of social marginalization that reflects a broader stigmatizing view of obesity.1 These have been accompanied by research looking at depression,2 psychiatric disorder,3,4 and quality of life.5,6 However, as with biomedical risk, the psychological distress associated with childhood obesity shows variation within and between studies. Any deficit in low self-esteem, for example, depends on children's gender, age, degree of obesity, and whether they are drawn from clinical or community samples.7
A specific problem lies with the conceptualization and measurement of self-esteem.8 Unidimensional or global self-esteem fails to capture the main sources from which self-worth is derived, in particular, body esteem or perception of physical appearance. Harter's9 perceived self-competence approach is a domain view of competence, and the questionnaire assessment provides a profile of self-rated competence in several domains and an assessment of global self-worth. Using this measure, reduced physical appearance and athletic competence have been observed in obese preadolescent girls,10 with social competence and global self-worth also lowered in some studies of obese teenagers.11,12
The method of eliciting information is also very different from that of most self-esteem assessments. Each question presents respondents with 2 contrasting statements, for example, "Some kids find it hard to make friends BUT Other kids find it pretty easy to make friends." This format permits an analysis of the number of respondents who identify with the positive or negative statement relevant to that domain. So, it is possible to evaluate the proportion of respondents in any group with low perceived self-competency, in addition to their mean score, something unreported previously in the child obesity literature.
The value of this approach, therefore, lies in its potential to address some of the weaknesses of previous research. For example, it is argued that even when low self-esteem is detected, the scores rarely fall outside of reference ranges, suggesting that it has little functional or clinical significance.13 Moreover, there is no indication of the proportion of obese children who have low self-esteem or whether there are particularly vulnerable subgroups who are being overlooked. It is also of note that most previous research has drawn on small clinical or modest community samples. There is clearly a need for evidence on the relationship between obesity and self-esteem from a large, representative community sample.
Accordingly, the present study aimed to examine the perceived self-competence of overweight and obese preadolescent children, across several competence domains and by prevalence of low competence, in a large representative community sample. Given that body dissatisfaction and low body esteem have been shown to account for higher levels of depression and low self-esteem, respectively,1,14 the possible mediation of low self-worth by body dissatisfaction was investigated. It was hypothesized that obese children would have lower self-competence in some but not all of the domains. Low appearance and athletic competence would be common to both obese girls and boys. In addition, body dissatisfaction would account for some but not all of the relationship between obesity and low perceived self-competence.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Materials
Body Height and Weight
Height was measured using a stadiometer with spirit level. Weight was measured using electronic scales, the participants wearing their school uniform without shoes and bulky outer clothing (eg, jumpers or jackets). Weight and height were measured to the nearest 0.1 kg and 0.5 cm, respectively. Children were classified into weight groups using the age standardized 2000 Centers for Disease Control and Prevention weight for height charts (underweight: 
5th percentile; normal weight: >5th and <85th percentile; overweight: 
85th <95th percentile; obese: 95th percentile).15
Self-Esteem
The Self-perception Profile for Children has 36 items and assesses perceived competence in 5 domains (scholastic competence, social acceptance, athletic competence, physical appearance, and behavioral conduct), together with global self-worth.16 The questionnaire was developed for children aged 8–14 years and has been shown to have good internal reliability with 
ranging from.73 to.84. The questions are in an alternative response format, as described earlier. The subscales have 6 questions, each scored 1–4, with 1 and 2 indicating low-perceived competence and 3 and 4 reflecting a high-perceived competence. Low domain competence was defined as a mean scale score of 2.0.
Body Shape Perception
Gender specific, 9-figure scales ranging from thin to fat body shapes, adapted from drawings used with children aged 6–10 years,17 investigated body shape perception. The validity and reliability of outline drawings for assessing current and ideal body figures and determining body dissatisfaction for preadolescents and young adolescents are considered good (test-retest reliability: 0.87 and 0.83).18,19 Participants made 2 choices: current perceived body shape and ideal body shape. Body dissatisfaction was the difference between the 2 so that a negative value was a thinner ideal than current shape.
Socioeconomic Status
Socioeconomic status (SES) was determined from children's home postcode or suburb using National Census data collected in 1996.20 The Urban Index of Relative Socio-Economic Advantage was used in the analysis. This variable is a general socioeconomic index, summarizing information related to the economic and social characteristics of families and households, as well as personal education qualifications and occupations. Scores were slightly negatively skewed, indicating more participants from higher SES backgrounds.
Procedure
On the day of testing, a standardized introduction to the study was read out to participants. Questionnaires were distributed to children in their class groups. Emphasis was placed on working alone, not discussing answers, reporting honestly and accurately, and that it was not a test. Explanations and examples were used to demonstrate how to answer each of the questionnaires. Participants were encouraged to ask questions if they did not understand. Height and weight were measured in private after completion of the questionnaire. Where possible, questionnaires were checked for completeness.
Data Analysis
Two-way analysis of variance or multivariate analysis of variance was used to test differences between groups (main factors: gender and weight group, with age and SES included as covariates), with Student-Neuman-Keuls posthoc tests. Body dissatisfaction outliers were checked and excluded if z>3. The perceived self-competency scores were converted to categorical data so that a score 2.0 was designated low competency. Relative risk was used to compare the proportion of students with low competency using the normal weight group as the reference group. Sobel's test for mediation determined whether body dissatisfaction mediated the relationship between BMI and perceived self-competence.21 The data were analyzed using SPSS 11.5 (SPSS Inc, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The impact on perceived self-competence in obese girls was more profound. They differed significantly from their normal weight peers in social acceptance, athletic competence, physical appearance, and global self-worth. The gender by weight category interactions reflected the lower perceived competencies of obese girls compared with obese boys. Indeed, the obese girls scored significantly lower than the obese boys in social acceptance, physical appearance, and global self-worth (all P < .05). Overweight girls were intermediate in their perceived competence, being significantly lower than normal weight peers but significantly higher than obese peers in athletic competence, physical appearance, and global self-worth.
The proportion of children and the relative risk of low perceived domain competency (scoring 2) are summarized in Tables 2 and 3. Obese boys and girls were between 2 and 4 times more likely than their normal weight counterparts to report low competence in athletics, physical appearance, and global self-worth. A third of obese boys and 63% of obese girls had low competency in the domain of physical appearance. Ten and 20%, respectively, had low global self-worth. Obese girls were also 2 times more likely to have low perceived social acceptance. Heightened risk of low global self-worth was present in overweight girls but not overweight boys. In contrast, underweight children had an elevated risk of low competence in only 1 domain, namely, boy's perceived athletic competence.
|
|
Body dissatisfaction varied by weight category (F1,2710 = 44.89; P < .001). Figure 1 shows that only the underweight children desired an ideal body shape bigger than their perceived current body shape. Boys and girls in all of the other weight groups expressed a desire for thinness. Of the obese children, 91% of boys and 96% of girls chose an ideal body shape that was thinner than their current perception.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Given this pattern of association, it would be as wrong to make the generalization that all obese children have low self-esteem as it would be to dismiss the increased risk of distress as a "myth of psychological maladjustment."13 Two issues are worth considering here. Why some domains of competence and not others, and why some obese children and not others?
Several studies with preadolescent or young adolescent children have noted that athletic competence and physical appearance are scored low by obese children.10–12 Low contentment with physical appearance should not be a surprise given previous descriptions of low body esteem in obese teenagers14 and the ubiquity of body dissatisfaction at this and older ages.22 Similarly, low athletic competence, perceived and actual, has been described in overweight children at this age.23 Most overweight children do not differ from normal weight peers in object-control skills, such as controlling an implement or ball, but are poorer in locomotor skills, including running, hopping, and jumping. The mechanics of moving a larger body may be a contributing factor, as may be the less-than-positive attitudes of some teachers. There is also a potential embarrassment factor, enhanced if performance is poor in the eyes of peers.
The greater risk of perceived low social acceptance, albeit only in girls, again might be expected, especially given the stigma associated with obesity and the social rejection attached to ratings of fat body shapes.24,25 Accounts of the social marginalization of overweight adolescents26 and fat teasing by peers27 also give credibility to this aspect of self-perception. In contrast, there is very little evidence of obesity having an impact on early academic achievement in mainstream schooling or of an association with poor behavior at this age.
An associated observation is that treatments that result in weight loss for obese children also show an improvement in the domains that best separate obese and normal weight participants. Accordingly, physical appearance and athletic competence, as well as global self-worth, significantly improved at the end of a summer weight loss camp12 and an inpatient treatment program.28 However, this psychological rehabilitation was not complete, because the domain scores were still below those observed in normal weight adolescents.
The question of why some but not all obese children have low perceived competency is more difficult to address. Although it is plausible that a greater risk of social trauma, such as being the victim of bullying29 or fat teasing,27 may make some obese children more vulnerable to low self-esteem, there is little direct causal evidence. However, experiences of shame (defined as self-reported experiences of being humiliated or degraded by others) in conjunction with parental separation or unemployment have been found to account for an observed relationship between BMI and depression in older adolescents.30 Social factors and family support are important to understanding vulnerability to child psychopathology. They are also integral to conceptualizations of resilience.31 It is argued that children who show strong resilience have access to protective or mediating factors in 3 broad areas: within themselves, in their families, and within the communities in which they live.32 Resilience in the context of child obesity remains to be explored. However, there is evidence that a supportive family environment may be important. For example, it has been noted in Mexico that parents value child fatness as a sign of health, and food treats are indicative of parental caring.33 It is interesting, therefore, that obese junior school-aged children from affluent Mexican families were indistinguishable from their normal weight peers in self-esteem and other psychological or social outcomes.
As hypothesized, body dissatisfaction accounted for some, but not all, of the relationship between BMI and low competence. The disparity between current and preferred body shape was generally a better mediator for boys than for girls. This suggests that the observed low self-competence of obese girls at least is based on more than a desire to be thinner than they currently perceive themselves. It also implies that reducing body dissatisfaction, even if that were possible, would not extinguish the reduced competence and low self-worth seen in these children.
There are some limitations to these findings. First, the overall participation rate was 55%. It is possible, therefore, that there were differences in the participation of children from different weight groups. However, comparing the proportion of obese children with Australian national survey data shows some similarity. In the present study, 9.6% of boys and 7.0% of girls were obese. The 1995 National Nutrition Survey found 4.7% of 7- to 15-year-old boys and 5.5% of girls were obese.34 Second, the present study was limited to a relatively narrow age range. Past research reviews suggest that the relationship between obesity and low self-esteem strengthens during adolescence.6,7 However, given that self-esteem declines during adolescence regardless of weight, it is difficult to say how the risk of obesity-associated low self-esteem would change. Third, this study says nothing about possible ethnic differences in the relationship between child obesity and self-esteem, despite drawing on participants from a diversity of backgrounds. This would require either a much larger or targeted sample.
The value of this research is in the identification of components of low self-esteem associated with child obesity and an indication of the risk and proportion of children affected. Much has been made of the cardiovascular risk associated with child obesity. Freedman et al,35 for example, have shown that children over the 95th BMI percentile are between 2 and 4.5 times more likely than normal weight peers to have raised total cholesterol and systolic and diastolic blood pressure indicative of hypertension. The present study has shown a similar level of risk of low perceived competence. Presenting the biomedical and psychological risks alongside each other should strengthen the case for properly organized and funded interventions. Some working estimates of the numbers affected should also be of help in resource planning and organization of treatment service delivery.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Address correspondence to Andrew J. Hill, PhD, Academic Unit of Psychiatry and Behavioural Sciences, School of Medicine, University of Leeds, 15 Hyde Terrace, Leeds LS2 9LT, United Kingdom. E-mail: a.j.hill{at}leeds.ac.uk
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
