- Email: [email protected]
Brief report: Young people at risk for eating disorders in Southeast Brazil
ARTICLE IN PRESS
Journal of Adolescence Journal of Adolescence 29 (2006) 313–317 www.elsevier.com/locate/jado
Brief report: Young people at risk for eating disorders in Southeast Brazil Tatiana Moyaa,, Bacy Fleitlich-Bilyka, Robert Goodmanb a Childhood and Adolescence Eating Disorders’ Attending, Teaching and Research Project (PROTAD), Bulimia and Eating Disorders Outpatient Program (AMBULIM), Institute of Psychiatry, University of Sao Paulo, R. Dr. Ovı´dio Pires de Campos, sem nu´mero, Pinheiros, Sa˜o Paulo, SP, CEP: 05403-010, Brazil b Department of Child and Adolescent Psychiatry, Institute of Psychiatry, King’s College London, De Crespigny Park, Box Number P085, London SE5 8AF, UK
Abstract A representative sample of 7–14-year-old young people in southeast Brazil ðN ¼ 1251Þ was assessed using standardized parent and youth interviews, thereby identifying an ‘at-risk’ group of young people who met one or more DSM-IV criteria for anorexia and/or bulimia nervosa. These young people were compared with an age and gender matched comparison group for body mass index (BMI) and socio-economic status (SES). The prevalence of young people at risk for eating disorders was 1.4% (higher in females and rising with age). ‘At-risk’ individuals did not differ from controls in BMI but were of higher SES. In Brazil, the link between symptoms of eating disorders and higher SES is not just a referral artefact but is evident in a representative community sample. This might reflect a stronger preference for thinness among more westernized social groups. r 2005 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved. Keywords: Eating disorders; Survey; Children; Adolescents; Social class; Body mass index
Corresponding author. Av. das Ame´ricas, no. 1.155, sala 1711, Barra da Tijuca, Rio de Janeiro, RJ, CEP: 22631-
000, Brazil. Tel.: +0055 21 2429 7651; fax: +0055 21 2439 9063. E-mail addresses: [email protected] (T. Moya), [email protected] (B. Fleitlich-Bilyk), [email protected] (R. Goodman). 0140-1971/$30.00 r 2005 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.adolescence.2005.05.006
ARTICLE IN PRESS 314
T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
Introduction There are very few studies of children and adolescents at risk of eating disorders in developing countries. Studies of risk factors for these disorders carried out in several countries around the world showed well-established risk factors such as female gender and the presence of negative selfevaluations (Jacobi, Hayward, de Zawaan, Kraemer, & Agras, 2004) but the relevance of other factors, such as body mass index (BMI) and socio-economic status (SES), remains controversial. To the best of our knowledge, there have been no epidemiological studies of children and adolescents at risk of eating disorders in Latin America, perhaps reflecting a lack of funding and validated measures. The present study uses data from a large representative sample of Brazilian 7–14-year olds to investigate the association of BMI and SES with symptoms of eating disorders, comparing ‘at-risk’ individuals with an age and gender matched comparison group.
Materials and methods The data were collected as part of an epidemiological survey carried out in a single district in southeast Brazil in the year 2000 (Fleitlich-Bilyk & Goodman, 2004). This survey investigated the prevalence of DSM-IV disorders in a representative community sample of 7–14-year-old Brazilian schoolchildren ðN ¼ 1251Þ, obtained in a two-stage procedure that involved random sampling of both free and private schools, followed by random sampling of pupils within schools (FleitlichBilyk & Goodman, 2004). Psychiatric symptoms were assessed using the Development and Well-Being Assessment (DAWBA), a structured interview administered by trained lay interviewers who also recorded verbatim accounts of any reported problems. The measure has been validated both in Britain and Brazil (Fleitlich-Bilyk & Goodman, 2004; Goodman, Ford, Richards, Gatward, & Meltzer, 2000). For children aged 7–10 years, only parents were interviewed; for 11–14-year olds, both parents and young people were interviewed. In the course of the interview, respondents were asked just one structured question related to eating disorders: the question for 11–14-year olds was ‘‘Do you have concerns about being thin or dieting?’’ and there was a similar question for parents. When the answer to these questions was ‘‘yes’’, more detailed information about the concerns was obtained using open-ended probes covering the nature, frequency, severity, chronicity and impact of symptoms. The answers to these open-ended questions were transcribed verbatim by the interviewers, capturing parents’ comments such as ‘‘She is obsessed with dieting. She deliberately doesn’t eat because she wants to get thinner. I have to check if she atey’’ and young people’s comments such as ‘‘I’m on a diet where I eat only fruits to get thinner very fast and I feel unwell, dizzy and I have to go to bed. It was at its worst yesterday because I didn’t have either breakfast or lunchy’’ A total of 1040 young people were both weighed and measured as part of the study, representing 83% of the total sample. The remaining 211 young people were not weighed or measured by the interviewers due to pressure of time or non-availability of the scales and stadiometer—the non-participants had not selectively dropped out because of concern about their weight. Height and weight were available for 78% (91/117) of those individuals where parents or young people reported concerns about dieting or weight, as compared with 84% (949/1134) of the
ARTICLE IN PRESS T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
315
remainder; this small difference was not significant (continuity-adjusted chi-square ¼ 2.24, 1 df, p40:1). SES was identified using a standardized Brazilian measure of affluence and parental educational (Associacao Nacional de Empresas de Pesquisa (ANEP), 2000). In response to the screening question, concerns about thinness or dieting were reported by at least one informant in the case of 117 young people. After reviewing the answers to subsequent open-ended questions on this group, we identified 18 young people who were ‘at risk’ of an eating disorder on the basis that they fulfilled one or more of the DSM-IV criteria for anorexia or bulimia nervosa, and experienced significant distress or social impairment as a result. This review was made blind to BMI and SES by one of the authors (TM), who is a psychiatrist with considerable experience of eating disorders in childhood and adolescence. Since the study was an investigation of the association between ‘at-risk’ status and BMI, we did not include low BMI as an indicator of risk status. When the accounts of parents and youths were discordant, the details in the verbatim transcript were reviewed to decide which informant’s account was more convincing. The ‘at-risk’ group was compared with the rest of the sample for analyses of age and gender. For each of the 18 young people in the ‘at-risk’ group, five age and gender matched ‘controls’ ðN ¼ 90Þ were chosen where neither parent nor the young person had described any concerns about dieting or weight. The most exact age matches were selected until a total of five was reached; when more than five young people were equivalently good matches, random selection was used to reduce the number to five.
Results The participation rate was 82% for private schools (237/288) and 83% for free schools (1014/ 1216), a non-significant difference. The mean age of the children of the survey was 11.0 years and girls made up 47% of the total sample (589/1251). The 18 ‘at-risk’ individuals (meeting at least one of the DSM-IV criteria for anorexia or bulimia nervosa, with resultant distress or social impairment) represent a prevalence of 1.4% (18/1251). ‘At-risk’ status varied by gender, being 2.2% for girls (13/589) and 0.8% for boys (5/662) (continuity-adjusted chi-square value ¼ 3.67, 1 df, p ¼ 0:056). The proportion of ‘at-risk’ individuals rose markedly with age (Fig. 1). Comparing the ‘at-risk’ and comparison groups, ‘at-risk’ individuals were more likely to be of higher SES than age and gender matched controls (Table 1). The two groups did not differ significantly on BMI scores: the mean (S.D.) for BMI measured in kg/m2 was 19.21 (2.65) for the ‘at-risk’ individuals, as compared with 19.16 (3.61) for the controls (t ¼ 0:052, 100 df, p ¼ 0:96).
Discussion In an epidemiological study of a large and representative sample from the southeast of Brazil, 1.4% of 7–14-year olds were estimated to be at risk for eating disorders. To the best of our
ARTICLE IN PRESS 316
T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
Percentage of children at risk (%)
3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 7-8
9-10 11-12 Age range (years)
12-14
Fig. 1. Variation with age in the proportion of ‘at-risk’ individuals.
Table 1 Socio-economic status of ‘at-risk’ individuals and their age and gender matched controls Proportion ðNÞ in each social class
At risk ðN ¼ 18Þ Controls ðN ¼ 90Þ
1 (lowest)
2
3
4
5 (highest)
22% (4) 33% (30)
33% (6) 36% (32)
6% (1) 20% (18)
22% (4) 6% (5)
17% (3) 6% (5)
p ¼ 0:04, chi-squared for trend.
knowledge, this is the first epidemiological study in Latin America of the prevalence of young people at risk for eating disorders. Most studies in developed countries find no association between SES and risk for eating disorders (Gardner, Stark, Friedman, & Jackson, 2000; Jones, Bennett, Olmsted, Lawson, & Rodin, 2001; Miotto, Coppi, Frezza, & Preti, 2003; Robinson et al., 1996), or find increased risk with lower SES (Miller, Verhegge, Miller, & Pumariega, 1999; Thomas, James, & Bachmann, 2002). By contrast, we found increased risk with higher SES. Our results are similar to the findings of Al-Subaie (2000) from another developing country, Saudi Arabia, where schoolgirls from higher social classes scored higher on the Eating Disorder Inventory—Drive for Thinness subscale. Al-Subaie attributed this finding to westernization, noting that many highly educated Saudis obtained some of their education in the West. Our findings suggest a similar effect in Brazil, namely a greater preference for thinness among young people from more affluent and probably more westernized families. Speculatively, developing countries might pass from a state of uniform non-westernization (with a low preference for thinness across the social spectrum) through a transient state of partial westernization (with greater preference for thinness in the most affluent social groups) to a state of uniform westernization (with a high preference for thinness across the social spectrum). If this is so, then Brazil might be moving from partial to uniform westernization in its attitudes to diet and body shape. Our study showed that just over half of the young people at risk came from the two lowest social classes. Even in very deprived shanty towns,
ARTICLE IN PRESS T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
317
almost every household in Brazil has a television set and therefore youths from different socioeconomic backgrounds will be exposed to similar media portrayals of the desirability of thinness. There were no statistically significant differences in BMI scores between risk and comparison groups. One plausible explanation is that ‘at-risk’ behaviours could be triggered by the young person initially being overweight, such that BMI subsequently falls from the high to the normal range. If this is confirmed by future studies, a single measure of BMI would not be a good way to screen Brazilian youth for their risk of eating disorder—serial BMIs or specific questions about the symptoms of eating disorders would seem more appropriate. A limitation of this study is that in the absence of validated diagnostic measures for eating disorders in Brazil, we relied on screening questions followed by clinical evaluation of open-ended comments. This may have led us to underestimate the number of young people at risk, particularly as a result of under-reporting of bulimic symptoms. While data on height and weight were only available for 83% of the sample, the missing data were determined by practical constraints and were not due to young people who were concerned about their weight refusing to be measured. Acknowledgements The work reported in this paper was supported by a project grant from the Wellcome Trust. References Al-Subaie, A. S. (2000). Some correlates of dieting behavior in Saudi schoolgirls. International Journal of Eating Disorders, 28, 242–246. Associacao Nacional de Empresas de Pesquisa (ANEP). (2000). Criterio de Classificacao Economica Brasil [Text file] URL: http://www.anep.org.br/codigosguias/CCEB.pdf. Fleitlich-Bilyk, B., & Goodman, R. (2004). The prevalence of child psychiatric disorders in Southeast Brazil. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 727–734. Gardner, R. M., Stark, K., Friedman, B. N., & Jackson, N. A. (2000). Predictors of eating disorder scores in children ages 6 through 14. A longitudinal study. Journal of Psychosomatic Research, 49, 199–205. Goodman, R., Ford, T., Richards, H., Gatward, R., & Meltzer, H. (2000). The Development and Well-Being Assessment: Description and initial validation of an integrated assessment of child and adolescent psychopathology. Journal of Child Psychology and Psychiatry, 41, 645–655. Jacobi, C., Hayward, C., de Zawaan, M., Kraemer, H. C., & Agras, W. S. (2004). Coming to terms with risk factors for eating disorders: Application of risk terminology and suggestions for a general taxonomy. Psychological Bulletin, 130, 19–65. Jones, J. M., Bennett, S., Olmsted, M. P., Lawson, M. L., & Rodin, G. (2001). Disorders eating attitudes and behaviours in teenaged girls: A school-based study. Canadian Medical Association Journal, 165, 547–552. Miller, M. N., Verhegge, R., Miller, B. E., & Pumariega, A. J. (1999). Assessment of risk of eating disorders among adolescents in Appalachia. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 437–443. Miotto, P., Coppi, M., Frezza, M., & Preti, A. (2003). The spectrum of eating disorders: Prevalence in an area of Northeast Italy. Psychiatry Research, 119, 145–154. Robinson, T. N., Killen, J. D., Litt, I. F., Hammer, L. D., Wilson, D. M., Haydel, K. F., Hayward, C., & Taylor, C. B. (1996). Ethnicity and body dissatisfaction: Are Hispanic and Asian girls at increased risk for eating disorders? Journal of Adolescent Health, 19, 384–393. Thomas, C. L., James, A. C., & Bachmann, M. O. (2002). Eating attitudes in English secondary school students: Influences os ethnicity, gender, mood, and social class. International Journal of Eating Disorders, 31, 92–96.
Journal of Adolescence Journal of Adolescence 29 (2006) 313–317 www.elsevier.com/locate/jado
Brief report: Young people at risk for eating disorders in Southeast Brazil Tatiana Moyaa,, Bacy Fleitlich-Bilyka, Robert Goodmanb a Childhood and Adolescence Eating Disorders’ Attending, Teaching and Research Project (PROTAD), Bulimia and Eating Disorders Outpatient Program (AMBULIM), Institute of Psychiatry, University of Sao Paulo, R. Dr. Ovı´dio Pires de Campos, sem nu´mero, Pinheiros, Sa˜o Paulo, SP, CEP: 05403-010, Brazil b Department of Child and Adolescent Psychiatry, Institute of Psychiatry, King’s College London, De Crespigny Park, Box Number P085, London SE5 8AF, UK
Abstract A representative sample of 7–14-year-old young people in southeast Brazil ðN ¼ 1251Þ was assessed using standardized parent and youth interviews, thereby identifying an ‘at-risk’ group of young people who met one or more DSM-IV criteria for anorexia and/or bulimia nervosa. These young people were compared with an age and gender matched comparison group for body mass index (BMI) and socio-economic status (SES). The prevalence of young people at risk for eating disorders was 1.4% (higher in females and rising with age). ‘At-risk’ individuals did not differ from controls in BMI but were of higher SES. In Brazil, the link between symptoms of eating disorders and higher SES is not just a referral artefact but is evident in a representative community sample. This might reflect a stronger preference for thinness among more westernized social groups. r 2005 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved. Keywords: Eating disorders; Survey; Children; Adolescents; Social class; Body mass index
Corresponding author. Av. das Ame´ricas, no. 1.155, sala 1711, Barra da Tijuca, Rio de Janeiro, RJ, CEP: 22631-
000, Brazil. Tel.: +0055 21 2429 7651; fax: +0055 21 2439 9063. E-mail addresses: [email protected] (T. Moya), [email protected] (B. Fleitlich-Bilyk), [email protected] (R. Goodman). 0140-1971/$30.00 r 2005 The Association for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.adolescence.2005.05.006
ARTICLE IN PRESS 314
T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
Introduction There are very few studies of children and adolescents at risk of eating disorders in developing countries. Studies of risk factors for these disorders carried out in several countries around the world showed well-established risk factors such as female gender and the presence of negative selfevaluations (Jacobi, Hayward, de Zawaan, Kraemer, & Agras, 2004) but the relevance of other factors, such as body mass index (BMI) and socio-economic status (SES), remains controversial. To the best of our knowledge, there have been no epidemiological studies of children and adolescents at risk of eating disorders in Latin America, perhaps reflecting a lack of funding and validated measures. The present study uses data from a large representative sample of Brazilian 7–14-year olds to investigate the association of BMI and SES with symptoms of eating disorders, comparing ‘at-risk’ individuals with an age and gender matched comparison group.
Materials and methods The data were collected as part of an epidemiological survey carried out in a single district in southeast Brazil in the year 2000 (Fleitlich-Bilyk & Goodman, 2004). This survey investigated the prevalence of DSM-IV disorders in a representative community sample of 7–14-year-old Brazilian schoolchildren ðN ¼ 1251Þ, obtained in a two-stage procedure that involved random sampling of both free and private schools, followed by random sampling of pupils within schools (FleitlichBilyk & Goodman, 2004). Psychiatric symptoms were assessed using the Development and Well-Being Assessment (DAWBA), a structured interview administered by trained lay interviewers who also recorded verbatim accounts of any reported problems. The measure has been validated both in Britain and Brazil (Fleitlich-Bilyk & Goodman, 2004; Goodman, Ford, Richards, Gatward, & Meltzer, 2000). For children aged 7–10 years, only parents were interviewed; for 11–14-year olds, both parents and young people were interviewed. In the course of the interview, respondents were asked just one structured question related to eating disorders: the question for 11–14-year olds was ‘‘Do you have concerns about being thin or dieting?’’ and there was a similar question for parents. When the answer to these questions was ‘‘yes’’, more detailed information about the concerns was obtained using open-ended probes covering the nature, frequency, severity, chronicity and impact of symptoms. The answers to these open-ended questions were transcribed verbatim by the interviewers, capturing parents’ comments such as ‘‘She is obsessed with dieting. She deliberately doesn’t eat because she wants to get thinner. I have to check if she atey’’ and young people’s comments such as ‘‘I’m on a diet where I eat only fruits to get thinner very fast and I feel unwell, dizzy and I have to go to bed. It was at its worst yesterday because I didn’t have either breakfast or lunchy’’ A total of 1040 young people were both weighed and measured as part of the study, representing 83% of the total sample. The remaining 211 young people were not weighed or measured by the interviewers due to pressure of time or non-availability of the scales and stadiometer—the non-participants had not selectively dropped out because of concern about their weight. Height and weight were available for 78% (91/117) of those individuals where parents or young people reported concerns about dieting or weight, as compared with 84% (949/1134) of the
ARTICLE IN PRESS T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
315
remainder; this small difference was not significant (continuity-adjusted chi-square ¼ 2.24, 1 df, p40:1). SES was identified using a standardized Brazilian measure of affluence and parental educational (Associacao Nacional de Empresas de Pesquisa (ANEP), 2000). In response to the screening question, concerns about thinness or dieting were reported by at least one informant in the case of 117 young people. After reviewing the answers to subsequent open-ended questions on this group, we identified 18 young people who were ‘at risk’ of an eating disorder on the basis that they fulfilled one or more of the DSM-IV criteria for anorexia or bulimia nervosa, and experienced significant distress or social impairment as a result. This review was made blind to BMI and SES by one of the authors (TM), who is a psychiatrist with considerable experience of eating disorders in childhood and adolescence. Since the study was an investigation of the association between ‘at-risk’ status and BMI, we did not include low BMI as an indicator of risk status. When the accounts of parents and youths were discordant, the details in the verbatim transcript were reviewed to decide which informant’s account was more convincing. The ‘at-risk’ group was compared with the rest of the sample for analyses of age and gender. For each of the 18 young people in the ‘at-risk’ group, five age and gender matched ‘controls’ ðN ¼ 90Þ were chosen where neither parent nor the young person had described any concerns about dieting or weight. The most exact age matches were selected until a total of five was reached; when more than five young people were equivalently good matches, random selection was used to reduce the number to five.
Results The participation rate was 82% for private schools (237/288) and 83% for free schools (1014/ 1216), a non-significant difference. The mean age of the children of the survey was 11.0 years and girls made up 47% of the total sample (589/1251). The 18 ‘at-risk’ individuals (meeting at least one of the DSM-IV criteria for anorexia or bulimia nervosa, with resultant distress or social impairment) represent a prevalence of 1.4% (18/1251). ‘At-risk’ status varied by gender, being 2.2% for girls (13/589) and 0.8% for boys (5/662) (continuity-adjusted chi-square value ¼ 3.67, 1 df, p ¼ 0:056). The proportion of ‘at-risk’ individuals rose markedly with age (Fig. 1). Comparing the ‘at-risk’ and comparison groups, ‘at-risk’ individuals were more likely to be of higher SES than age and gender matched controls (Table 1). The two groups did not differ significantly on BMI scores: the mean (S.D.) for BMI measured in kg/m2 was 19.21 (2.65) for the ‘at-risk’ individuals, as compared with 19.16 (3.61) for the controls (t ¼ 0:052, 100 df, p ¼ 0:96).
Discussion In an epidemiological study of a large and representative sample from the southeast of Brazil, 1.4% of 7–14-year olds were estimated to be at risk for eating disorders. To the best of our
ARTICLE IN PRESS 316
T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
Percentage of children at risk (%)
3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 7-8
9-10 11-12 Age range (years)
12-14
Fig. 1. Variation with age in the proportion of ‘at-risk’ individuals.
Table 1 Socio-economic status of ‘at-risk’ individuals and their age and gender matched controls Proportion ðNÞ in each social class
At risk ðN ¼ 18Þ Controls ðN ¼ 90Þ
1 (lowest)
2
3
4
5 (highest)
22% (4) 33% (30)
33% (6) 36% (32)
6% (1) 20% (18)
22% (4) 6% (5)
17% (3) 6% (5)
p ¼ 0:04, chi-squared for trend.
knowledge, this is the first epidemiological study in Latin America of the prevalence of young people at risk for eating disorders. Most studies in developed countries find no association between SES and risk for eating disorders (Gardner, Stark, Friedman, & Jackson, 2000; Jones, Bennett, Olmsted, Lawson, & Rodin, 2001; Miotto, Coppi, Frezza, & Preti, 2003; Robinson et al., 1996), or find increased risk with lower SES (Miller, Verhegge, Miller, & Pumariega, 1999; Thomas, James, & Bachmann, 2002). By contrast, we found increased risk with higher SES. Our results are similar to the findings of Al-Subaie (2000) from another developing country, Saudi Arabia, where schoolgirls from higher social classes scored higher on the Eating Disorder Inventory—Drive for Thinness subscale. Al-Subaie attributed this finding to westernization, noting that many highly educated Saudis obtained some of their education in the West. Our findings suggest a similar effect in Brazil, namely a greater preference for thinness among young people from more affluent and probably more westernized families. Speculatively, developing countries might pass from a state of uniform non-westernization (with a low preference for thinness across the social spectrum) through a transient state of partial westernization (with greater preference for thinness in the most affluent social groups) to a state of uniform westernization (with a high preference for thinness across the social spectrum). If this is so, then Brazil might be moving from partial to uniform westernization in its attitudes to diet and body shape. Our study showed that just over half of the young people at risk came from the two lowest social classes. Even in very deprived shanty towns,
ARTICLE IN PRESS T. Moya et al. / Journal of Adolescence 29 (2006) 313–317
317
almost every household in Brazil has a television set and therefore youths from different socioeconomic backgrounds will be exposed to similar media portrayals of the desirability of thinness. There were no statistically significant differences in BMI scores between risk and comparison groups. One plausible explanation is that ‘at-risk’ behaviours could be triggered by the young person initially being overweight, such that BMI subsequently falls from the high to the normal range. If this is confirmed by future studies, a single measure of BMI would not be a good way to screen Brazilian youth for their risk of eating disorder—serial BMIs or specific questions about the symptoms of eating disorders would seem more appropriate. A limitation of this study is that in the absence of validated diagnostic measures for eating disorders in Brazil, we relied on screening questions followed by clinical evaluation of open-ended comments. This may have led us to underestimate the number of young people at risk, particularly as a result of under-reporting of bulimic symptoms. While data on height and weight were only available for 83% of the sample, the missing data were determined by practical constraints and were not due to young people who were concerned about their weight refusing to be measured. Acknowledgements The work reported in this paper was supported by a project grant from the Wellcome Trust. References Al-Subaie, A. S. (2000). Some correlates of dieting behavior in Saudi schoolgirls. International Journal of Eating Disorders, 28, 242–246. Associacao Nacional de Empresas de Pesquisa (ANEP). (2000). Criterio de Classificacao Economica Brasil [Text file] URL: http://www.anep.org.br/codigosguias/CCEB.pdf. Fleitlich-Bilyk, B., & Goodman, R. (2004). The prevalence of child psychiatric disorders in Southeast Brazil. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 727–734. Gardner, R. M., Stark, K., Friedman, B. N., & Jackson, N. A. (2000). Predictors of eating disorder scores in children ages 6 through 14. A longitudinal study. Journal of Psychosomatic Research, 49, 199–205. Goodman, R., Ford, T., Richards, H., Gatward, R., & Meltzer, H. (2000). The Development and Well-Being Assessment: Description and initial validation of an integrated assessment of child and adolescent psychopathology. Journal of Child Psychology and Psychiatry, 41, 645–655. Jacobi, C., Hayward, C., de Zawaan, M., Kraemer, H. C., & Agras, W. S. (2004). Coming to terms with risk factors for eating disorders: Application of risk terminology and suggestions for a general taxonomy. Psychological Bulletin, 130, 19–65. Jones, J. M., Bennett, S., Olmsted, M. P., Lawson, M. L., & Rodin, G. (2001). Disorders eating attitudes and behaviours in teenaged girls: A school-based study. Canadian Medical Association Journal, 165, 547–552. Miller, M. N., Verhegge, R., Miller, B. E., & Pumariega, A. J. (1999). Assessment of risk of eating disorders among adolescents in Appalachia. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 437–443. Miotto, P., Coppi, M., Frezza, M., & Preti, A. (2003). The spectrum of eating disorders: Prevalence in an area of Northeast Italy. Psychiatry Research, 119, 145–154. Robinson, T. N., Killen, J. D., Litt, I. F., Hammer, L. D., Wilson, D. M., Haydel, K. F., Hayward, C., & Taylor, C. B. (1996). Ethnicity and body dissatisfaction: Are Hispanic and Asian girls at increased risk for eating disorders? Journal of Adolescent Health, 19, 384–393. Thomas, C. L., James, A. C., & Bachmann, M. O. (2002). Eating attitudes in English secondary school students: Influences os ethnicity, gender, mood, and social class. International Journal of Eating Disorders, 31, 92–96.