Validity and reproducibility of Escala de Evaluación da Insatisfación Corporal para Adolescentes
Validación y reproducibilidad de la Escala de Evaluación de la Insatisfacción Corporal para Adolescentes
Maria Aparecida ContiI; Betzabeth SlaterII; Maria do Rosário Dias de Oliveira LatorreIII
IInstituto de Psiquiatria. Faculdade de Medicina. Universidade de São Paulo (USP). São Paulo, SP, Brasil
IIDepartamento de Nutrição. Faculdade de Saúde Pública (FSP). USP. São Paulo, SP, Brasil
IIIDepartamento de Epidemiologia. FSP-USP. São Paulo, SP, Brasil
OBJECTIVE: To validate a body dissatisfaction scale for adolescents.
METHODS: The study included 386 female and male adolescents aged ten to 17 years enrolled in a private elementary and middle school in the city of São Bernardo do Campo, Southeastern Brazil, in 2006. "Escala de Evaluación da Insatisfación Corporal para Adolescentes" (body dissatisfaction scale for adolescents) was translated and culturally adapted. The Portuguese instrument was evaluated for internal consistency using Cronbach's alpha, factor analysis with Varimax rotation, discriminant validity by comparing score means according to nutritional status (low weight, normal weight, and at risk of overweight and obesity) using the Kruskal-Wallis test. Concurrent validity was assessed using Spearman's rank correlation coefficient between scores and body mass index, waist-hip ratio and waist circumference. Reproducibility was evaluated using Wilcoxon test, and intraclass correlation coefficient.
RESULTS: The translated and back-translated scale showed good agreement with the original one. The translated scale had good internal consistency in all subgroups studied (males and females in early and intermediate adolescence) and was able to discriminate adolescents according to their nutritional status. In the concurrent analysis, all three measures were correlated, except for males in early adolescence. Its reproducibility was ascertained.
CONCLUSIONS: The "Escala de Evaluación da Insatisfación Corporal para Adolescentes" was successfully translated into Portuguese and adapted to the Brazilian background and showed good results. It is recommended for the evaluation of the attitudinal component of body image in adolescents.
Descriptors: Adolescent. Self Concept. Body Image. Scales. Translations. Validity of Tests. Reproducibility of Results.
OBJETIVO: Validar escala de insatisfacción corporal para adolescentes.
MÉTODOS: Participaron del estudio 386 adolescentes, de ambos sexos, entre diez y 17 años de edad, de una escuela particular de enseñanza fundamental y media, de São Bernardo do Campo, Sureste de Brasil, en 2006. Fueron realizadas traducción y adaptación cultural de la "Escala de Evaluación de Insatisfacción Corporal para Adolescentes" para el portugués. Fueron evaluados consistencia interna por medio del coeficiente alfa de Cronbach, análisis factorial por el método Varimax y validez discriminante por las diferencias entre promedios de estado nutricional, utilizándose la prueba de Kruskal-Wallis. En la validación concurrente, se calculó el coeficiente de correlación de Spearman entre la escala y el índice de masa corporal, la razón circunferencia cuadril y la circunferencia de la cintura. Para reproducibilidad, fueron utilizados la prueba de Wilcoxon, el coeficiente de correlación intra-clase.
RESULTADOS: La escala traducida no presentó discordancias significativas con la original. La escala presentó consistencia interna satisfactoria para todos los subgrupos estudiados (fases inicial y final de adolescencia, ambos sexos) y fue capaz de discriminar los adolescentes según el estado nutricional. En el análisis concurrente, las tres medidas corporales fueron correlacionadas, excepto adolescentes del sexo masculino en fase inicial, y su reproducibilidad fue confirmada.
CONCLUSIONES: La Escala de Evaluación de la Insatisfacción Corporal para Adolescentes esta traducida para el portugués y presentó resultados satisfactorios, siendo recomendada para evaluación del aspecto de la actitud de la imagen corporal de adolescentes.
Descriptores: Adolescente. Autoimagen. Imagen Corporal. Escalas. Traducción (Producto). Validez de las Pruebas. Reproducibilidad de Resultados.
Body image is the mental image that the individual has of his own body, surrounded by its immediate sensations and experiences.21 It is a type of "mental picture" that the person makes of his or her physical appearance and the attitudes and feelings in relation to that image. It is also the result of various experiences accumulated during their lifetime, which are mediated by the central nervous system.3 Cash & Pruzinsky3 describe it as a multidimensional construct with at least two independent modalities: perception, related to an estimation of body size, and attitude, related to affection and cognition.
Because it is a comprehensive construct, various techniques for measuring it are used, such as the drawing of the human figure, the scale of bodily areas, figure silhouettes, photographs, distorted mirror and video-tapes and visual bodily estimations, in addition to questionnaires and satisfaction scales.2 According to the research target, all techniques have their methodological advantages and weaknesses.
In the case of the application of the technique of the drawing of a human figure, the difficulty lies in the absence of points of reference for partially or fully judging the individual measurements collected, thus making it difficult to determine the degree of perceptive distortion.1 Body area and silhouette scales, on the other hand, are widely used because of the practicality of application and correction. However, their power to provide information about the theme is limited, because they only evaluate the degree of satisfaction in relation to image and specific body areas.23 To apply the photography technique, a lot of investment in equipment and research materials is necessary, in addition to it demanding individual attention to the subject.2 Distorting mirrors and videotapes reveal the difficulty of adjusting parts to the whole, once it is difficult to alter a single part of the body independently of the others.1 The visual body estimation technique demands a large investment in research material, it takes a long time to apply it and is individualized, thus requiring a suitable location.1
In population studies questionnaires and scales are preferable due to their ease of application, the possibility of large samples and, if necessary, the application of other instruments.17
Among the questionnaires available in literature the Escala de Evaluación de Insatisfación Corporal para Adolescentes12 (EEICA) [Body Shape Dissatisfaction Assessment Scale for Adolescents] still has no version in Portuguese and its validity and the possibility of it being reproduced to meet the Brazilian reality have not been studied. To develop the Spanish version of the scale, six stages were covered, involving 569 young people of both sexes, enrolled in high schools in Navarra, Spain. In the first stage, 42 assessment items were proposed, based on the clinical work of the authors and on a bibliographic review. In the following stage, four specialists from the area of eating disorders and body image refined the initial items and eliminated seven of them. Then, the questions that had a correlation coefficient less than 0.20 were removed during the exploratory factor analysis phase. In the following stage, the reliability and reproducibility of the scale were analyzed and Cronbach's Alpha values obtained; they varied between 0.66 and 0.93 and the intra-class correlation coefficient was 0.90.12
Translated and culturally adapted instruments, which have been validated for the Brazilian reality, guarantee that the information obtained is representative of the group being surveyed. These instruments can also be applied in epidemiological, clinical and health prevention and promotion contexts.
The EEICA assesses body dissatisfaction in young males and females between 12 and 19 years old. The advantages of this scale relative to others relates to how easy it is to apply and correct and to the fact that it obtains information relating to the frequency of behavior related to body care, body perception and family and social influences.
The objective of this study was to validate the body dissatisfaction scale for adolescents.
The EEICA was translated in the second half of 2005 after prior authorization from the authors of the original scale. Initially, the scale questions were translated from Spanish into Portuguese by three researchers with experience in validating instruments. The researchers discussed the discrepancies in the translations and arrived at a consensus on the version in Portuguese. This version was then translated back into Spanish by a professor who is a nutrition specialist and whose native language is Spanish. Finally, the Spanish version was compared with the original text to define the final version of the scale in Portuguese (Attachment).
The EEICA comprises 32 questions for completion by the subject in the form of a Likert scale and has six response categories: 1 - never; 2 - hardly ever; 3 - sometimes; 4 - often; 5 - almost always; 6 - always. The score is calculated in the following way: questions with a positive direction (questions 1-5, 7-9, 11-17, 19, 20, 22-26, 28, 30 and 31) receive 0 for 'never', 'hardly ever' and 'sometimes' responses; 1 for 'often' and 2 for 'almost always' 3 for 'always' responses. Those questions with a negative direction (questions 6, 10, 18, 21, 27, 29 and 32) receive 0 for 'always', 'almost always' and 'often' responses, 1 for 'sometimes', 2 for 'hardly ever' and 3 for 'never' responses. The score is calculated from the sum of responses and varies from 0 to 96 points; the higher the number of points the greater the young person's dissatisfaction with his/her body.
The sample comprised all adolescents regularly enrolled in a private elementary and high school in the city of São Bernardo do Campo, Southeastern Brazil, in 2006, who were between ten and 17 years old. The questionnaires of 80 of the 466 young people enrolled could not be considered. Eight adolescents said they did not want to take part because they were apparently overweight. There were five young people whose parents refused authorization for them to take part and 65 did not bring back the free and informed consent form, making it impossible for them to participate. One adolescent was excluded because she was pregnant, which would have compromised the assessment of the anthropometric measurements. One adolescent who was taking psychiatric medication responded to the questionnaire with the help of a researcher, but was excluded from the analysis. In the end, 386 young people were included in the study.
Data collection took place in March 2006. A pre-test was carried out in August, the questionnaires were applied at moment 1 (test) and moment 2 (re-test) and anthropometric measurements were taken between September and December 2006.
The questionnaire was collectively applied in the classroom (test) and then the adolescent was sent to the gymnasium for the anthropometric measurements to be taken. At the end of this stage, a second interview (re-test) was booked for three weeks later.12,23
To check the degree of understanding after applying the questionnaire, the young people were asked the question: "Did you understand what was being asked in this scale?".12 The replies were of the Likert-scale type: 0 - I didn't understand anything; 1 - I understood a little; 2 - I understood more or less; 3 - I understood almost everything, but I had some doubts; 4 - I understood perfectly well and I have no doubts.
The anthropometric assessment was carried out by the author of this research (MAC). Electronic platform scales with capacity to weigh up to 150 kg in 100 gr sub-divisions were used to establish body weight; the adolescents wore light clothing and no shoes, according to the methodology proposed by Gordon et al.7 Height was measured using a stadiometer (SECA) fixed to the wall, with a scale in millimeters (mm). The adolescents were asked to stand with their backs to the wall and their heels, calves, buttocks and shoulders touching it. The head was positioned in the Frankfort plane, in accordance with the methodology proposed by Gordon et al.7 Weight and height were measured twice and the average of the values was considered. A tape measure was used to measure the waist and hips; it was firmly applied around the waist, at the narrowest part of the trunk, and around the buttocks.
Calculation of the body mass index (BMI) was done using the equation: BMI = weight (kg)/height (m)2.
Classification of the state of nutrition of the adolescents was carried out according to the recommendation of the World Health Organization (WHO)25 for this age band: low weight <5th percentile; normal weight = between 5th percentile and < 85th percentile; rick of overweight and obesity > 85th percentile.
A descriptive statistical analysis was carried out and average, standard deviation and minimum and maximum values were calculated. Psychometric properties were checked by internal consistency (Cronbach's Alpha coefficient), factor analysis, discriminant and convergent validity and reproducibility.
In the factor analysis, items were submitted to analysis to find out the joint variation pattern and the variance of each factor, using orthogonal rotation (Varimax method). In the discriminant validation process, the averages of the EEICA of the three groups were compared: low weight, normal weight and risk of overweight and obesity25 using the Kruskal-Wallis one-way analysis of variance. It was expected that adolescents at risk of being overweight and obese would express greater dissatisfaction when compared with the others. For convergent validity, Spearman's correlation coefficient was calculated between the scale score and the BMI, waist measurement (WM) and waist-hip ratio (WHR). It was expected that the greater the BMI, WM or WHR the greater would be the degree of dissatisfaction.
Reproducibility was assessed by comparing the averages of the scale scores at the two research moments (test and re-test), using Wilcoxon's test, the intra-class correlation coefficient (rintraclasse) and the Bland-Altman graph. For analysis of the time needed to conclude the scale the average time taken by the adolescent for understanding and completing it was recorded and analyzed by comparing the percentages and average and standard deviation values.
Validation and reproducibility analyses were carried out in accordance with the phases of adolescence: initial (between 10 and 14) and intermediary (between 15 and 17).25
To input data and for consistency and descriptive analysis of the same the Epi Info version 6.04 for DOS was used. For other analyses the statistical package, SPSS version 10.0 was used and for Bland & Altman graphs MedCalc was used.
The study was approved by the Ethics and Research Committee of the Faculdade de Saúde Pública of the Universidade de São Paulo.
When translating into Portuguese, it was necessary to adjust the vocabulary in three questions, verbal conjugations in two and adaptations with regard to the use of pronouns in two.
For example, the word "convertirte", was translated into Portuguese as "tornar-se" [become], "transformar-se" [be transformed]. The expression "has llegado", which literally translated would be "tinha chegado" [had arrived] was translated as "chegou" [arrived], to more faithfully express the content and verb tense of the question.
There was a need to make adaptations, like the use of pronouns, because in Spanish it is normal to use "teu" [your - familiar] and in Portuguese the non-colloquial form uses the pronoun "seu" [your - formal]. In the translation, therefore, it was decided to adopt the non-colloquial use of pronouns in order to get closer to the translated content.
The expression "esthetic surgery" was translated as "plastic surgery", because it refers to the alteration or change of some area of the body. Another word that presented a certain cultural difference was "pastilla". Although initially translated as "drug", it was decided to use the word "comprimido" [tablet], because in the Brazilian culture, drug refers to any type of medication, which would include medicine, pills and tablets.
Another word that needed adapting on the original scale was the word "fisico" [physique], which was translated into Portuguese as "body shape", because it more closely expressed the original content. Its literal translation would refer to a limited meaning of the content of the question.
There were 386 participants in initial adolescence: 29.3% (113) were male and 33.2% (128) female. In the intermediary adolescence category, 16.8% (65) and 20.7% (80) were male and female, respectively (Table 1)
As to weight, height and BMI the average values observed and their respective standard deviations [SD] corresponded to 57.5 kg (SD=14.4); 161.1 cm (SD=10.0) and 22.0 (SD=4.1).
The average time taken to apply the scale was between 4.5 min and 5.1 min and the average for verbal understanding varied between 3.4 and 3.7 for the four sub-groups studied.
The questions with the greatest frequency of "always" replies, thus inferring greater body dissatisfaction, were: Question 5 (How often do you think you would like to have more will power to control what you eat?) with 13.2%, Question 13 (How often are you afraid of losing control and becoming fat?) with 11.9% and Question 2 (How often do you think you would see yourself in a better light if you could wear smaller-sized clothes?) with 10.9% (Table 2).
The questions that had the greatest frequency of "never" replies, thus inferring greater body dissatisfaction, were: Question 32 (How often do you feel you would like to be fatter?) with 75.4%; Question 27 (How often do you feel thin?) with 27.5%; Question 21 (How often do you feel very good when trying on clothes before buying them, particularly pants?) with 21.0% and Question 6 (How often do you think that your body shape is the shape currently considered attractive?) with 19.7%.
The scale varied from 1 to 87, with an average of 20.9 points (SD= 14.2) and a median of 16 points.
In the internal consistency analysis, Cronbach's Alpha coefficient (á) varied from 0.72 to 0.93 (Table 3) and the factor analysis revealed that the questions explain 54.7% of the total variability of the data. The factors resulting from the factor analysis were submitted to internal consistency analysis and varied from 0.63 to 0.85, which indicates moderate to high consistency, respectively. In the discriminant validation process, there were statistically significant differences in the averages of the scores between the three sub-groups studied. In the convergent validation analysis, statistically significant correlations were recorded between the four sub-groups, except for male adolescents in the initial phase of adolescence between the WHR and WM, although the correlation value for WHR was at the significance limit (Table 3).
In relation to reproducibility, there was no significantly statistical difference in the average scores when moment 1 (test) and moment 2 (re-test) were considered. The correlation coefficients between the scores of moments 1 and 2 were significant, varying from 0.64 (p<0.001) to 0.91 (p<0.001). The Figure shows that the EEICA was randomly well distributed around zero, with few points outside the limit for the four sub-groups.
Factor analysis confirmed that the questions explain 54.7% of the variance of the instrument, and, in the original study, this value was 50.3%. In the discriminant validation analysis the EEICA was able to discriminate between the three groups (thinness, normal weight and risk of overweight and obesity) in all the sub-groups analyzed.
The convergent validity for the three measurements between the males and females in the initial and intermediary phases of adolescence was confirmed. The correlation values were between 0.21 (p=0.005) and 0.44 (p<0.001); 0.23 (p<0.001) and 0.44 (p<0.001); 0.28 (p<0.001) and 0.39 (p<0.001) for BMI, WHR and WM, respectively.
Cash,5 Thompson & Gray23 and Thompson24 underline the importance of carrying out studies like this one before applying instruments developed in other countries.
The translation was facilitated by the similarity between Spanish and Portuguese, which have retained close characteristics in relation to spelling and pronunciation of its signs and phonemes. Even so, cultural differences were observed that required adaptations in the translation process in order for it to be as approximate and equivalent to the content taken from the original scale as possible.14
The EEICA assesses attitudinal aspects of the body image of adolescents. Three types of validation analysis were developed: convergent, construct and discriminant. In convergent validation, the EEICA was compared with three validated instruments adapted to the Spanish reality: the "Eating Disorder Inventory"9 (EDI) body dissatisfaction sub-scale, the "Body Shape Questionnaire"6 (BSQ) and the "Eating Attitudes Test"8 (EAT). In the construct validation, the points trend was determined and the consistent factors were grouped using factor analysis. In the discriminant validation, two groups that had been selected according to the EAT8 points criterion and the EDI9 sub-scale were compared. In this study the same criteria were not used, because these instruments (EDI, EAT and BSQ) have not been validated for a population of young Brazilians of both sexes.
It is usual to apply weight measurements and the BMI10,17,23 in validation studies, because assessment studies of the distortion of body image in children and adolescents that would make instruments and measurements available for comparison purposes are rare. This being so, because there was no gold standard test by which to assess body image, it was assumed that overweight and obese young people mean a greater degree of dissatisfaction and this was the criterion for assessing both discriminant and convergent validity.
Most of the studies and scientific production that focus on infant-juvenile development have prioritized the degree of body satisfaction, associating it with aspects like age, sex, nutritional state, quality of family interaction, social pressure and others. There is agreement that young overweight and obsess people of both sexes are more dissatisfied than their peers.16,20
In this study the internal consistency of the EEICA was between 0.72 (males in the intermediary phase of adolescence) and 0.93 (females in the initial phase of adolescence). Guillén Grima et al12 recorded Cronbach's Alpha values of 0.86, 0.93 and 0.92 for the, respectively, 12-13 years, 14-16 and 17-19 years female age groups. For males, the values were 0.74 and 0.66 for the 12-13 and 17-19 years age groups, respectively. The internal consistency of this study is similar to the values in the original study.
In convergent validation, Guillén Grima et al,12 recorded a variation in the indices for EAT (r=0.67, 0.52, 0.70); BSQ (r=0.81, 0.51, 0.88) and EDI (r=0.81, 0.51, 0.83) for the whole sample, for both male and female adolescents, respectively. It can be seen that these were higher than those encountered in the present study.
This confirms, therefore, the discriminant validity of all groups studied in the EEICA. Convergent validity was confirmed with the three correlation measurements (BMI, WHR and WM for both sexes in the initial and intermediary phases of adolescence, except for males in the initial phase of adolescence in the correlation with the WM.
In relation to reproducibility, between moments 1 and 2, the EEICA was the same in comparison with the averages for all sub-groups surveyed and the intra-class correlation varied from 0.64 (males in the intermediary adolescence) to 0.91 (females in the initial adolescence). Guillén Grima et al12 found a value of 0.84 for the whole sample and 0.86 and 0.50 for males and females, respectively. Therefore, it can be concluded that the EEICA proved to be stable for all the sub-groups surveyed.
The average lowest time for applying the EEICA was 4.5 minutes and the average verbal understanding of the scale was greater than 3.4 (maximum possible = 4) for all sub-groups, thus constituting a rapidly applied and easily understood instrument.
Body dissatisfaction is a fact of life for many adolescents. Levine & Smolak15 infer that in developed countries approximately 40%-70% of all young females are dissatisfied with their bodies, with more than 50% of them aspiring to be thin. For young males there are no such accurate data, but it is known that there is dissatisfaction and aspirations are divided between getting thinner and increasing muscle mass.16 In Brazilian studies, it is possible to see that young females are more dissatisfied than young males.5,19
Thought should be given as to how to establish when body dissatisfaction starts being a problem for young people so that an unusual or pathological degree of body dissatisfaction can be distinguished.22
Regarding the conduct of young people, given their dissatisfaction with their bodies, approximately 40% of North American females and 25% of North American males start dieting in adolescence.18 Dissatisfied young people also prove to be more susceptible to the development of eating pathologies, like nervous anorexia and bulimia or obesity.13
Standardized instruments, like the one presented here, constitute useful and reliable tools for assessing the body dissatisfaction of young people within the epidemiological, clinical, health prevention or promotion context.
The EEICA, as translated and adapted into Portuguese, was internally consistent and confirmed discriminant and convergent validities and reproducibility for all sub-groups. Satisfactory values were obtained for understanding and interval of time needed to complete the questionnaire. The conclusion is that the EEICA is an instrument recommended for assessing the attitudinal aspect of the body image for adolescents.
1. Banfield SS, McCabe MP. An evaluation of the construct of body image. Adolescence. 2002;37(146):373-93.
2. Bergstrom E, Stenlund H, Svedjehall B. Assessment of body perception among swedish adolescent and young adults. J Adolesc Health. 2000;26(1):70-5. DOI: 10.1016/S1054-139X(99)00058-0
3. Cash TF, Pruzinsky T., orgs. Body image: a handbook of theory, research, and clinical practice. New York: Guilford; 2002.
4. Cash TF. Body image: past, present, and the future. Body Image. 2004;1(1):1-5. DOI: 10.1016/S1740-1445(03)00011-1
5. Conti MA, Frutuoso MF, Gambardella AMD. Excesso de peso e insatisfação corporal em adolescentes. Rev Nutr. 2005;18(4):491-7. DOI: 10.1590/S1415-52732005000400005
6. Cooper Z, Cooper PJ, Fairburn CG. The validity of the eating disorder examination and its subscales. Br J Psychiatry. 1989;154:807-12. DOI: 10.1192/bjp.154.6.807
7. Gordon CC, Chumlea WC, Roche AF. Stature, recumbent length, and weight. In: Lohman TG, Roche AF, Martorell R. Anthropometric standardization reference manual. Champaing: Human Kinetics; 1988. p.327-399
8. Garner DM, Garfinkel PE. The eating attitudes test:an index of symptoms of anorexia nervosa. Psychol Med. 1979;9(2):270-3.
9. Garner D, Olmstead M, Polivy J. Development and validation of a multidimensional eating disorder inventory for anorexia and bulimia. Int J Eating Disord. 1983;2:15-34. DOI: 10.1002/1098-108X(198321)2:2<15::AID-EAT2260020203>3.0.CO;2-6
10. Gardner RM. Body image assessment of children. In: Cash TF, Pruzinsky T, orgs. Body image: a handbook of theory, research, and clinical practice. New York: Guilford, 2002. p.127-134.
11. Grassi-Oliveira R, Stein LM, Pezzi. Tradução e validação de conteúdo de versão em português de Childhood trauma questionnaire. Rev Saude Publica. 2006;40(2):249-55. DOI: 10.1590/S0034-89102006000200010
12. Guillén Grima F, Garrido Landívar E, Baile Ayensa JI. Desarrollo y validación de una escala de insatisfacción corporal para adolescentes. Med Clin (Barc). 2003;121(5):173-7.
13. Haines J, Neumark-Sztainer D. Prevention of obesity and eating disorders: a consideration of shared risk factors. Heath Educ Res. 2006;21(6):770-82. DOI: 10.1093/her/cyl094
14. Herdman M, Fox-Rushby J, Badia X. "Equivalence" and the translation and adaptation of health-related quality of life questionnaires. Qual Life Res. 1997;6(3):237-47. DOI: 10.1023/A:1026410721664
15. Levine MP, Smolak L. Body image development in adolescence. In: Cash, TF, Pruzinsky T, orgs. Body image: a handbook of theory, research, and clinical practice. New York: Guilford; 2004.
16. McCabe MP, Ricciardelli LA. Body image dissatisfaction among males across the lifespan: a review of past literature. J Psychosom Res. 2004;56(6):675-85. doi: 10.1016/S0022-3999(03)00129-6
17. Mendelson BK, Mendelson MJ, White DR. Body-esteem scale for adolescence and adult. J Pers Assess.. 2001;76(1):90-106. DOI: 10.1207/S15327752JPA7601_6
18. Nicholls D, Viner R. Eating disorders and weight problems. BMJ. 2005;330(7497): 950-3. DOI: 10.1136/bmj.330.7497.950
19. Pinheiro AP, Giugliani ERJ. Body dissatisfaction in Brazilian schoolchildren: prevalence and associated factors. Rev Saude Publica. 2007;40(3):489-96.
20. Richards MH, Petersen AC, Boxer AM, Albrecht R. Relation of weight to body image in pubertal girls and boys from two communities. Dev Psychol. 1990;26(2):313-21. DOI: 10.1037/0012-16126.96.36.1993
21. Schilder P. A Imagem do corpo: as energias construtivas da psique. São Paulo: Martins Fontes; 1981.
22. Smolak L. Body image in children and adolescent: where do we go from here? Body Image. 2004;1(1):15-28. DOI: 10.1016/S1740-1445(03)00008-1
23. Thompson MA, Gray JJ. Development and validation of a new body-image assessment scale. J Pers Assess. 1995;64(2):258-69. DOI: 10.1207/s15327752jpa6402_6
24. Thompson JK. The (mis)measurement of body image: ten strategies to improve assessment for applied and research purposes. Body Image. 2004;1(1):7-14. DOI: 10.1016/S1740-1445(03)00004-4
25. World Health Organization. Physical status: the use and interpretation of anthropometry. Geneva; 1995. (WHO Technical Reports Series, 854). Correspondence: Received: 10/11/2007 Article based on the doctoral thesis Conti MA, presented to the à Faculdade de Saúde Pública da Universidade de São Paulo in 2007. ANNEX
Maria Aparecida Conti
AMBULIM - Instituto de Psiquiatria
Faculdade de Medicina - USP
R. Dr. Ouvídio Pires de Campos, 785, 2 º andar
05403-010 São Paulo, SP, Brasil
Conti MA was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq; Processo 140097/2005-8; doctoral fellowship).
Article based on the doctoral thesis Conti MA, presented to the à Faculdade de Saúde Pública da Universidade de São Paulo in 2007.