Acta Pñdiatr 88: 998±1003. 1999
Mental health and psychosocial characteristics in adolescent obesity: a population-based case-control study ˚ man1 C Renman1, I Engstro¨m2, S-A Silfverdal1 and J A ¨ rebro Medical Centre Hospital; Psychiatric Research and Development Unit2, O ¨ rebro, Sweden Department of Paediatrics1, O
˚ man J. Mental health and psychosocial characteristics in Renman C, Engstro¨m I, Silfverdal S-A, A adolescent obesity: a population-based case-control study. Acta Pædiatr 1999; 88: 998–1003. Stockholm. ISSN 0803–5253 In this population-based study we compared self-esteem, social background, social and academic competence, behavioural problems and lifestyle in 58 obese adolescents (BMI 99.6th percentile or 30 kg/m2), aged 14–18 y, with 58 sex- and age-matched controls of normal weight. The instruments used were: I Think I Am, Youth Self Report and a lifestyle questionnaire. The obese group was on average, 40 kg heavier than the controls. The obese individuals rated themselves significantly lower in physical characteristics, but in all other aspects of self-esteem, mental health and social and academic competence there were no differences between the two groups. There were significant socioeconomic differences, with more obese adolescents living with only one parent and with the mothers in the obese group having, in general, lower education than those in the control group. This study confirms previous observations that obesity is associated with special socioeconomic conditions in youth, but that obese adolescents do not differ from their normalweight peers in other aspects of mental health. & Adolescents, obesity, self-esteem, social behaviour ¨ rebro Medical Centre Hospital, S-701 85 O ¨ rebro, Sweden J A˚man, Department of Paediatrics, O (Tel. 46 19 151000, fax. 46 19 187915, e-mail. [email protected]
The increasing prevalence of overweight and obesity among adults, adolescents and children is a disquieting international trend reported from different countries (1, 2). The body mass index (BMI) has increased most markedly in the heaviest part of the population (2). There is a common opinion that obesity in childhood has negative psychosocial consequences (3). Others have reported that overweight in adolescence and early adulthood is associated with impaired socioeconomic development, especially among women (4). The relationship between obesity, mental health and self-esteem in children and adolescents is not fully understood. In an extensive review of the literature, French et al. (5) concluded that cross-sectional studies showed a negative association between relative weight and body-esteem at the ages of 7–12 y, while the correlation with self-esteem was inconsistent. In adolescents, of ages 13–18 y, an inverse relationship was usually found between relative weight and both bodyand self-esteem (5). Previous prospective population-based studies have shown conflicting results regarding the relationship between body weight and self-esteem (4, 6, 7). In one of these, in which a 7-y prospective follow-up was carried Scandinavian University Press 1999. ISSN 0803-5253
out on a large sample of adolescents and young adults, no correlation was found between overweight and selfesteem (4). Another prospective study showed similar results (7). In a third study with a 3-y follow-up, an inverse relationship was found between BMI and selfesteem regarding physical appearance in both males and females and between BMI and global self-esteem in females (6). Prospectively, in females, self-esteem at baseline was also inversely related to BMI 3 y later, both with regard to physical appearance and social acceptance (6). In general, clinical hospital-based studies more often indicate negative associations between obesity and selfesteem in comparison with results from populationbased studies. This discrepancy could be due to methodological weakness regarding selection of patients in some of the hospital-based studies (5, 7). It might also suggest that there are subpopulations of obese children and those seeking help maybe more vulnerable to development of impaired self-esteem (7). The aim of the present study was to estimate the prevalence of severe obesity in a population- based cohort of adolescents and, furthermore, to study the self-esteem, mental health, social background and
Obesity and self-esteem in adolescents
ACTA PÆDIATR 88 (1999)
Table 1. Prevalence of obesity according to BMI among the complete population of adolescents, aged 14–18 y, attending comprehensive ¨ rebro. school and upper secondary school in the city of O BMI Comprehensive school Upper secondary school Total (n = 6.319)
99.6 percentile and/or BMI 30
2.9% 2.4% 2.5%
0.8% 1.0% 0.9%
0.8% 1.3% 1.2%
0.1% 0.4% 0.3%
0.9% 1.4% 1.2%
lifestyle in obese adolescents compared with an ageand sex-matched control group of adolescents of normal weight.
Patients and methods Study population A complete sample of adolescents, aged 14–18 y, from ¨ rebro, Sweden, was the public schools in the city of O studied with respect to weight and BMI. The Swedish national growth charts for weight and height (8) were used as the basic instrument to identify overweight, defined as >2 SD in weight related to height. Data were collected from the school health files. BMI defined as weight in kg/height in m2 was calculated. Severe obesity was defined as BMI 30 and/or BMI 99.6th percentile in accordance with recently published BMI reference curves for the UK (9). In the investigated population, 1.2% of the pupils (76/6.319) were identified as having severe obesity according to these criteria. Within this group, 0.3% of the pupils (17/6.319) had a BMI 35 (Table 1). Three pupils did not participate because they had moved from the area. Two pupils were excluded from the following questionnaire study, one due to mental retardation and another because of an exceptionally high muscle mass rather than obesity. The remaining 71 pupils were asked to participate in the study through a letter written to their families. The participation rate was 86% (43/50) in upper secondary school and 71% (15/21) in comprehensive school. Thus, 58 pupils (30 boys and 28 girls) with severe obesity agreed to participate. For each obese individual, one non-obese control pupil of the same age and sex in the parallel class of the same school was identified. In that class, the pupil who was born closest to the obese pupil was asked to participate through a letter to the family. The refusal rate was low among the control pupils and refusing controls were replaced by the pupil who was second closest to the obese pupil. The mean age was 16.9 y in both groups. The obese pupils were on average 40 kg heavier than the controls, with a mean weight of 100.7 14.0 kg and a BMI of 33.9 3.2. Physical characteristics of the two groups are shown in Table 2. The BMI of the 13 nonparticipating obese pupils was 32.3, which is similar to that of the participating obese pupils.
The study was approved by the Medical Ethics ¨ rebro County Council. Committee of O Methods The psychosocial assessment was made using three instruments concerning self-esteem, social competence, behavioural problems and lifestyle. All instruments were questionnaires, which were filled in individually at the offices of the school nurses. The pupils were informed by the school nurse that all data were going to be handled anonymously. Written instructions were given to the pupils in the questionnaires. All pupils completed the questionnaires during the school day. The first two questionnaires were standardized and validated for Swedish adolescents. The third questionnaire was constructed for this particular study in order to evaluate physical activities, eating habits and other aspects of lifestyle. The study was performed in spring 1997, during a time span of 2 mo in both obese pupils and controls. I Think I Am (ITIA). This questionnaire was developed in Sweden (10) for measuring self-esteem in children and adolescents. The instrument yields a total score for self-esteem, which is divided into five subscales measuring the child’s physical characteristics, skills and talents, mental characteristics, relationships with parents and relationships with others. Each scale consists of equally divided positive and negative statements ordered by chance. Youth Self Report (YSR). This checklist, derived from the CBCL (Child Behaviour Checklist) provides a standardized self-description of social competence and behaviour problems. The social competence scale measures sports, hobbies, activities, friends and acaTable 2. Physical characteristics in the studied population of 58 obese adolescents in comparison with 58 adolescents who were not overweight (mean SD).
Age (y) Weight (kg) Height (m) BMI (kg/m2)
Obese group (n = 58)
Non-obese group (n = 58)
16.9 1.1 100.7 14.0 1.72 0.08 33.9 3.2
16.9 1.0 60.6 9.0 1.72 0.08 20.5 2.0
0.930 4 h weekly >6 h weekly
Obese group (%) (n = 58)
Non-obese group (%) (n = 58)
69 74 45 34
76 83 56 40
0.406 0.240 0.230 0.564
21 24 12 56 26 3.4 5.2 5.2
23 25 10 29 25 3.6 11 3.6
0.638 0.830 0.695 0.003 0.996 0.972 0.260 0.677
29 40 24 3.4
28 50 33 12
0.517 0.263 0.304 0.083
girls were found in any of the YSR or ITIA scales. Neither were any gender obesity status interactions observed for any of the demographic or psychosocial variables. In the Spearman rank correlation test, we found no correlation between BMI and mental health measured according to the methods and scales described above, except for one subscale in the ITIA. Not unexpectedly, there was a negative correlation between BMI and selfevaluated physical characteristics (r = ÿ0.387; p < 0.0001).
Discussion In this population-based study of 58 obese adolescents who were 40 kg overweight compared with closely matched controls, we found that their obesity had only a minor correlation with self-reported self-esteem. As expected, the obese pupils were not satisfied with their body and their physical abilities. Adolescence is the period with the highest risk for developing disturbances in body image (12). On the other hand, no overall indications of impairments were detected according to variables measuring other aspects of mental health. These findings confirm previous observations made in population-based studies (4, 7). Clinical studies of severely obese adults undergoing surgery for their obesity have also shown that there are no single personality impairments or indications of psychopathol-
ogy in the obese population (12, 13). Thus, our study does not provide any indication that screening procedures are needed in the obese population to identify individuals with impaired mental health or decreased self-esteem. The mental health of children seeking medical care for obesity may, however, be impaired and should therefore be assessed thoroughly. We found some social differences between the two groups, with a higher proportion of the obese adolescents living with only one parent, having mothers with a lower level of education and more often having family members with smoking habits. These findings suggest that our obese adolescents live in families with a lower socioeconomic standard and less education. We do not think that these differences are dependent on the selection procedure of the control group, which was carefully identified in the same school as the obese pupils. Previous authors (4, 14, 15) have concluded that obesity and socioeconomic status have a complex inverse relationship and the present observations indicate that this also holds true in youth. Social factors thus seem to be correlated with the presence of adolescent obesity. The reason for the correlation with one parent is not clear from this study. It may be correlated with socioeconomic or emotional factors that enhances the risk for obesity, but to make any safe conclusions on this part, further and deeper studies have to be made. The purpose of the questions concerning eating habits, leisure activities and the use of drugs was to
C Renman et al.
ACTA PÆDIATR 88 (1999)
Table 6. Assessment of psychological symptoms from the Youth Self Report (YSR) in 58 obese adolescents in comparison with 58 adolescents who were not overweight (mean SD). Obese group n = 58 Internalized symptoms (total sum) Withdrawn Somatic complaints Anxious/depressed Social problems Thought problems Externalized symptoms (total sum) Attention problems Delinquent behaviour Aggressive behaviour
Non-obese group n = 58
Table 7. Assessment of self-esteem from I think I am (ITIA) in 58 obese adolescents in comparison with 58 adolescents who were not overweight (mean SD).
10.8 9.8 3.0 2.6 3.6 3.2 5.8 5.4 2.3 2.1 2.1 2.0
8.9 8.8 2.7 2.8 2.6 2.4 5.2 4.9 2.0 1.9 1.9 2.0
0.287 0.496 0.144 0.663 0.479 0.799
12.9 7.1 4.7 2.9 4.1 2.8 8.8 5.1
12.5 5.8 4.4 2.5 3.9 2.4 8.7 3.9
0.894 0.565 0.649 0.907
Physical characteristics Skills and talents Mental characteristics Relations with parents Relations with others ITIA total score
Obese group n = 58
Control group n = 58
6.1 8.3 9.7 7.9 11.2 9.3 15.7 10.0 12.6 7.7 52.5 33.6
3.7 9.1 9.3 7.6 13.5 9.4 15.9 9.7 13.4 6.9 64.5 33.3
p value 30 for men and >28.6 for women (19, 20). The WHO
definition of severe overweight or obesity in adulthood is BMI > 30 (21, 22). In childhood, BMI curves are helpful, since BMI increases with age and differs between boys and girls. BMI 99.6th percentile for age and sex is representative of the severely obese child. In adolescence, however, the BMI limit of 30 crosses the 99.6th percentile curve at the age of 15 y in boys and at the age of 13.3 y in girls (9). We therefore chose to define obesity as the BMI 30 and/or 99.6th percentile. We also found that the prevalence of severe obesity, defined as BMI 99.6th percentile, among our background population of adolescents was high compared with the reported figures for corresponding ages in the reference curves by Cole et al. (9). Our finding of a high prevalence of severe obesity is in line with previous reports regarding weight gain in adolescents and young adults (1, 2). In contrast to several other studies, our sample of adolescents was strictly population-based and we had a high rate of participation. The non-responders were not heavier than those who participated in the study. We therefore believe that the participating obese pupils are representative of the obese adolescents in this population. On the other hand, the sample size was relatively small and we cannot therefore exclude the possibility that tendencies to differences in some of the lifestyle variables might have become significant if the sample size had been larger. In conclusion, apart from dissatisfaction with their body and their physical abilities, the obese adolescents were comparable to matched controls in self-reported self-esteem, mental health and lifestyle habits. However, differences in socioeconomic conditions were observed between families with and without obese adolescents and this observation requires further investigation. The impaired social conditions might be of importance in adolescent obesity and therefore need to be taken into consideration if intervention strategies are planned.
¨ rebro for their Acknowledgements.—We thank all the school nurses in O invaluable help with the growth charts. This investigation was supported ¨ rebro County Council. by grants from the Research Committee of O
Obesity and self-esteem in adolescents
ACTA PÆDIATR 88 (1999)
References 1. Ro¨ssner S. Childhood obesity and adulthood consequences. Acta Paediatr 1998; 87: 1–5 2. Troiano RP, Flegal KM. Overweight children and adolescents: description, epidemiology and demographics. The causes and health consequences of obesity in children and adolescents. Pediatrics 1998; 101: 497–504 3. Dietz WH. Health consequences of obesity in youth: childhood predictors of adult disease. The causes and health consequences of obesity in children and adolescents. Pediatrics 1998; 101: 518–25 4. Gortmaker SL, Must A, Perrin JM, Sobol AM, Dietz AM, Dietz WH. Social and economic consequences of overweight in adolescence and young adulthood. N Engl J Med 1993; 329: 1008–12 5. French SA, Story M, Perry CL. Self-esteem and obesity in children and adolescents: a literature review. Obes Res 1995; 3: 479–90 6. French SA, Perry CL, Leon GR, Fulkerson JA. Self-esteem and change in body mass index over three years in a cohort of adolescents. Obes Res 1996; 4: 27–33 7. Rumpel C, Harris TB. The influence of weight of adolescent selfesteem. J Psychom Res 1994; 38: 547–56 8. Karlberg P, Taranger J, Engstro¨m I, Lichstenstein H, SvennbergRedgren I. The somatic development of children in a Swedish urban community. A prospective longitudinal study. Acta Paediatr Scand 1976; Suppl 258: 9. Cole TJ, Freeman JV, Preece MA. Body mass index reference curves for the UK, 1990. Arch Dis Child 1995; 73: 25–9 10. Ouvinen-Birgerstam. Jag tycker jag a¨r, manual. Psykologifo¨rlaget AB Stockholm, 1985 11. Achenbach TM. Manual for the Youth Self-Report and Profile.
14. 15. 16. 17. 18. 19. 20. 21. 22.
1991; Burlington, UT: University of Vermont Department of Psychiatry Wadden TA, Stunkard AJ. Psychological aspects of severe obesity. Am J Clin Nutr 1992; 55: 524–32 Wadden TA, Stunkard AJ. Psychopathology and obesity. Department of Psychiatry, University of Pennsylvania School of Medicine Philadelphia, Pennsylvania 19104–3246. Annals New York Academy of Sciences: 55–65 Stunkard AJ, Sørensen TIA. Obesity and socioeconomic status —a complex relation. N Engl J Med 1993; 329: 1036–7 Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of the literature. Psychol Bull 1989; 105: 260–75 Maffeis C, Zaffanello M, Schultz Y. Relationship between physical inactivity and adiposity in prepubertal boys. J Pediatr 1997; 131: 288–92 Dietz WH, Gortmaker SL. Do we fatten our children at the television set? Obesity and television viewing in children and adolescents. Pediatrics 1985; 75: 807–12 Bandini LG, Schoeller DA, Dietz WH. Energy expenditure in obese and non-obese adolescents. Pediatr Res 1990; 27: 198–203 Kuskowska-Wolf A, Bergstro¨m R. Trends in body mass index and prevalence of obesity in Swedish women 1980–89. J Epidemiol Commun Health 1993; 47: 195–9 Kuskowska-Wolf A, Bergstro¨m R. Trends in body mass index and prevalence of obesity in Swedish men 1980–89. J Epidemiol Commun Health 1993; 47: 103–8 Sjo¨stro¨m L. TV och stillasittande go¨r allt fler feta. La¨kartidningen 1996; 93: 167–70 (in Swedish) WHO International Obesity Task Force Report. Geneva: WHO, 1997.
Received Sept. 29, 1998; revisions received March 3, 1999 and Apr. 26, 1999; accepted May 6, 1999