Parental Factors, Mass Media Influences, and the Onset of Eating Disorders in a Prospective Population-Based Cohort
* Department of Epidemiology and Public Health, Medical School, University of Navarra, Pamplona, Spain
Department of Psychiatry, International University of Catalonia, Barcelona, Spain
Department of Psychiatry, University Clinic, University of Navarra, Pamplona, Spain
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective. To identify risk factors for eating disorders.
Methods. A community cohort study was conducted in Navarra, Spain. A region-wide representative sample of 2862 girls who were 12 to 21 years of age completed the Eating Attitudes Test (40-item version) and other questionnaires in 1997. Girls who scored high in the Eating Attitudes Test-40 were interviewed by a psychiatrist who applied Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria to diagnose prevalent cases of eating disorders. Girls who were free of any eating disorder in 1997 were reassessed after 18 months of follow-up using the same methods.
Results. Ninety new cases of eating disorders according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria were identified during the follow-up. In the multivariate logistic analysis, a higher risk of incident eating disorder was found for several exposures assessed at the beginning of follow-up, such as younger age, usually eating alone (odds ratio [OR]: 2.9; 95% confidence interval: 1.9–4.6), and frequently reading girls’ magazines or listening to radio programs (OR: 2.1; 1.2–3.8 for those most frequently using both media). No independent association was found for television viewing or socioeconomic status. A marital status of parents different from "being married" was associated with a significantly higher risk in the multivariate analysis (OR: 2.0; 1.1–3.5).
Conclusions. Our results support the role of mass media influences and parental marital status in the onset of eating disorders. The habit of eating alone should be considered as a warning sign of eating disorders.
Key Words: eating disorders • anorexia nervosa • bulimia nervosa • mass media • parental
Abbreviations: ED, eating disorder • EAT, Eating Attitudes Test • DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition • AN, anorexia nervosa • BN, bulimia nervosa • EDNOS, eating disorder not otherwise specified • BMI, body mass index • SE, standard error • OR, odds ratio • CI, confidence interval
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The prevalence of anorexia nervosa is approximately 0.5% and of bulimia nervosa approximately 2% among female adolescents and young women in Western countries.1 These prevalences and the incidence of eating disorders (EDs) are high among girls and young women compared with other chronic diseases that affect women of these ages. An appropriate knowledge of the main modifiable determinants of ED is needed to develop effective preventive interventions. However, the available information about the determinants of eating disorders is mainly based on case-control2–4 or cross-sectional studies. In these designs, the possibility that some of the identified risk factors may be attributable to differential recall of past events or to reverse causality (ie, early consequences mistakenly identified as supposed exposures) cannot be ruled out. As randomized trials would not be feasible or ethical for these diseases, the most reliable information should come from prospective cohort studies.
A previous prospective study that included 1947 adolescents of both genders identified that dieting and psychiatric morbidity were strongly associated with the onset of EDs.5 In addition to these factors, which in some cases could be early symptoms of disease,6 some parental7,8 and cultural9 influences are believed to be risk factors. Among these factors, some characteristics related to the influences of mass media have been more extensively studied with controversial results10,11 and the evidence is sparse. To our knowledge, only a previous prospective study has assessed the association between media exposure and the development of abnormal eating behaviors among adolescent girls.12 However, only the self-reported use of laxatives or vomiting to control weight was used as the outcome in that study. Moreover, the case definitions in most of the previous studies addressing other risk factors for EDs were not based on clinically accepted diagnostic criteria but rather on responses to different versions of the Eating Attitudes Test (EAT). Studies that assessed clinically diagnosed cases of ED included a series of cases in a clinical setting but usually did not use population screening to ascertain additional cases.13,14 To our knowledge, no previous epidemiologic study has reported an association between exposure to mass media and EDs in a population sample using accepted diagnostic criteria applied by a psychiatrist experienced in diagnosing these disorders.
We assessed the association between several parental, mass media, sociodemographic, and psychosocial influences and EDs, using cases of ED confirmed with Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria by experienced psychiatrists, in a large representative sample of the female population of 12- to 21-year-olds in Navarra, Spain.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Initial Selection of Participants in the Cohort
The present study constitutes the first prospective assessment of a cohort started in 1997. The methods of our study have been previously described.15,16 Briefly, a weighted, random, multistage sampling scheme was used to select a representative sample of the female adolescent population (12–21 years of age) of Navarra (an autonomous region in the north of Spain with approximately 500 000 inhabitants). Participation in the study was obtained for 39 of the 43 schools thus selected. The randomly selected classrooms of the 39 schools that agreed to enter the study included 3472 potential participants. A total of 610 girls did not complete the screening questionnaire; the main reason for not participating was school absenteeism. The initial cohort therefore constituted 2862 girls aged 12 to 21 years. Informed consent was obtained from the girls (if they were in college) or from their parents if they were younger than college students.
Measures
At baseline, a double-phase procedure was used to identify prevalent cases of ED. First, a screening phase was performed using the 40-item version of the EAT.17 This Spanish version has been previously validated.18 Then each possible case (every participant who scored >30 on the EAT-40) was confirmed using a semistructured interview performed by a psychiatrist who was experienced in EDs and applied DSM-IV criteria to diagnose anorexia nervosa (AN), bulimia nervosa (BN) or "eating disorder not otherwise specified" (EDNOS specifically defined in DSM-IV as "eating disorder not otherwise specified" category is for disorders of eating that do not meet the criteria for any specific ED). Only psychiatrist-confirmed diagnoses according to DSM-IV criteria were considered as prevalent cases. As we have reported elsewhere,15 we found 119 prevalent cases of ED (AN: 9 cases; BN: 22 cases; EDNOS: 88 cases) in the initial assessment of the cohort. Only 4 of them had been previously diagnosed (3 girls had received a diagnosis of AN, and 1 had received a diagnosis of BN). All prevalent cases were excluded from the follow-up analyses.
Parental, mass media exposures, and other sociodemographic and lifestyle characteristics were collected at baseline in self-administered questionnaires together with the EAT. A scale with 36 items19 was used to appraise self-esteem in 4 different domains: social, emotional, familial, and academic. This scale has been previously validated in a large sample of Spanish adolescents and is more comprehensive than the tool used by the majority of previous studies; the shorter Rosenberg’s scale contains only 10 or 15 items. In addition, the Rosenberg’s scale has not yet been validated in the Spanish population.
To assess the potential role of solitary eating, we inquired with a simple question in the baseline questionnaire whether the girls had the habit of eating alone. The exact wording of the question was, "Do you usually eat alone?" with only 2 options in the response (yes/no).
Body weight was measured using an electronic scale calibrated to the nearest 100 g. Height was measured with an electronic device calibrated to the nearest centimeter. Body mass index (BMI) was calculated as weight in kilograms divided by height in squared meters.
Follow-up
Participating girls who were free from EDs at baseline were contacted and reevaluated again after 18 months of follow-up using similar procedures. After also excluding those who had failed to provide reliable EAT questionnaires in 1997 (n = 11), we successfully followed up 2509 girls who completed the EAT-40 again (follow-up proportion: 92%). Among them, those who scored higher than 21 in this test (n = 446) were invited to be examined by a psychiatrist. This was possible for 88% of them (n = 394). Among them, we found 78 incident cases of ED. For the remaining 12% of girls who scored higher than 21 in the EAT-40 and we were unable to interview (n = 52), a panel of experienced psychiatrists used the answers to the EAT-40 and to the Eating Disorders Inventory,20 which was also applied to each participant, to ascertain the diagnosis of ED. Twelve new cases were found among them. In either interviews or expert panels, psychiatrists were blinded to mass media exposures and parental characteristics recorded in the baseline questionnaires.
Statistical Analysis
Odds ratios were used to estimate relative risks. We estimated their 95% confidence intervals using standard methods. The psychiatrist-confirmed diagnosis of ED according to DSM-IV criteria was used as the outcome. Multivariate logistic models were used to ascertain the independent association of each potential risk factor with incident ED, allowing to control for a wide set of confounders (age at baseline assessment, socioeconomic status, marital status of parents, self-esteem, and BMI). After univariate analyses, we followed the purposeful method for the selection of candidate variables.21 We did not adjust for dieting or excessive exercise at baseline because they are believed to be early signs or potential manifestations of ED.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ninety incident cases of psychiatrist-confirmed ED according to DSM-IV criteria were identified during the follow-up period. The mean EAT score was significantly higher among cases (mean: 39.4; standard error [SE]: 1.4) than among noncases (mean: 13.6; SE: 0.2). Cases presented a significantly higher BMI (mean: 22.2; SE: 0.3) than noncases (mean: 21.6; SE: 0.1). We did not find any association between the development of an ED and the number of siblings, birth order, previous psychiatric disorders in the family, or recent familial stressful events. More than 90% of girls who were scheduled for an interview could be evaluated directly by the psychiatrist, whereas the remaining smaller number of girls with high EAT scores had to be evaluated indirectly by examining their responses to the EAT questionnaire. A sensitivity analysis was performed to verify whether bias could have resulted from the diagnosis of some incident cases using such questionnaire items instead of an interview. The analyses performed showed no evidence of such bias, and subsequent analyses were consequently performed disregarding this difference in diagnostic procedure (data not shown).
The proportion of girls who developed some ED was higher among girls whose parents were separated, divorced, or widowed than among girls whose parents were married. The habit of solitary eating was present in 440 girls at baseline. Among them, we identified 35 incident cases of ED (8%). This proportion was 3 times higher than the incidence among those who did not usually eat meals alone, ie, they ate them with family members or other people.
Regarding the use of mass media, we found no differences in the incidence of EDs according to the overall number of hours the girls watched television during a typical week or when we separately analyzed the use of television during workdays or weekends. However, the linear trend approached statistical significance in the univariate analysis for a positive association with the number of hours of television watching during a typical weekend day (odds ratio [OR]: 1.05; 95% confidence interval [CI]: 0.99–1.11 for each 1-hour increase; P = .13). However, we found a statistically significant association showing that the incidence of ED increased in a stepwise mode with the time spent listening to radio programs (OR: 1.11; 95% CI: 1.03–1.20 for each 1-hour increase; P = .008). In addition, more frequently reading teen girls magazines (at least once a week) was also associated with a higher risk of developing ED (OR: 1.55; 95% CI: 1.01–2.37; P = .045).
We studied the association of several variables with the habit of eating alone, because they could be potential confounders of the association (Table 1). We found that solitary eating was more frequent in the 14- to 16-year stratum, in adolescents from lower socioeconomic levels, and among daughters of separated/divorced/widowed parents. We also found an apparent inverse association with self-esteem19: the higher the self-esteem, the lower the frequency of solitary eating. We did not observe any difference in the proportion of girls who reported solitary eating at baseline between those who heavily used mass media (radio, television, or magazines) and those who did not (data not shown). Therefore, there was a need to introduce the variables shown in Table 1 in multivariate models to control for confounding.
|
In the multivariate logistic regression analysis, after adjusting for socioeconomic status (3 categories), BMI (kg/m2), and self-esteem,19 we found the results shown in Table 2. In the column of multivariate analysis, all ORs are adjusted for the variables shown in the table in addition to BMI, self-esteem, and socioeconomic status. Solitary eating, marital status of parents different from being married, and a higher level of radio use were independent predictors for a higher risk of developing an ED. The highest magnitude of the association was found for the habit of solitary eating.
|
We decided to build a combined score to assess the joint effect of the 2 mass media for which we found associations with ED risk at least approaching statistical significance. We assigned 1 point to girls who were in the upper stratum of radio use (>1 hour/d) and 1 point to those who were in the upper stratum of girls’ magazine use (at least once weekly) and summed them, thus creating a single 3-category variable (0, 1 point, 2 points). We introduced this new categorical variable in a multivariate logistic model (Table 3). In addition to BMI, self-esteem, socioeconomic status, and the variables shown in the Table 3, we adjusted the estimates for previous existence of any ED in other members of the same family. We found a significantly higher risk (2.1-fold) for girls in the high category of using both mass media (radio and magazines). The linear trend for this variable (combined use of both mass media) was statistically significant (P = .015).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In an 18-month follow-up of 2509 girls who were initially free of any ED after a screening phase, we subsequently found a high incidence of new cases of ED, although most of them were only partial syndromes (EDNOS, according to the DSM-IV criteria). Of these, >80% presented characteristics more closely resembling bulimia than anorexia. We found that the habit of eating alone; a situation of divorce, separation, or widowing in parents; and a higher use of some mass media (not television) were associated with a higher risk of developing clinically defined ED. To our knowledge, this is the first prospective study to identify such associations.
Because of the apparent rise in incidence of ED and their clinical importance,22,23 the need to conduct prospective studies to assess their main risk factors has been highlighted very often.24–26 In the absence of effective treatment,23 the emphasis should be placed on prevention.1,12 Risk factor identification for planning effective preventive interventions remains essential.
Other prospective studies have been conducted recently.5,12,27–30 However, some of them have used only answers to questionnaires as the outcome12,29 but not interviews to identify new cases of clinically defined disease. Among the few prospective studies that used clinical criteria,5,28–30 the present study represents the largest sample size. Moreover, our cohort was based on a representative sample of girls who were 12 to 21 years of age at baseline, selected using a random sampling procedure with a whole Spanish region as the sampling frame. This design increases the external validity of our results.
Solitary eating was identified as a major risk factor for ED. Owing to the scarcity of methodologically sound prospective studies, little is known with certainty about this or other risk factors. This factor, nevertheless, has not been studied extensively. It is true that solitariness is common in other conditions such as depression or substance abuse. However, we did not identify any such conditions in our sample, and this enables us to suggest that eating alone was independently associated with ED in our study. Although perhaps it is not a specific predictor of ED, this finding is nevertheless useful for the purpose of "early detection or prevention" because it indicates that this, albeit nonspecific, distress can indeed lead to or be an early sign of a severe condition such as an ED. Among eating patterns, previous studies have consistently identified dieting as a powerful risk factor.5 Despite some criticisms about the actual possibility of reverse causation because dieting can be an early consequence of an underlying ED,6,31 these findings have led to a wide consensus supporting the theory that diet is a first step leading in the long run to full-blown forms of ED, which are the extreme of a continuum spectrum of disease.12,32 One of the earliest manifestations in this continuum, preceding dieting, could be the habit of eating alone. This finding, if confirmed by additional studies, may have implications for prevention. Parents and educators should be cautioned to avoid letting girls eat alone frequently, because this habit could be the environment where the ED develops.
We found that parental marital status was associated with the onset of ED. Therefore, our results provide support to previous evidences about the higher psychiatric risk imposed by family breakdowns and parental separation or divorce.7 Some consistency of our results with previous cross-sectional data26 reinforce the strength of a potential cause and effect relationship. A strong social support coming from a stable family has been very often advocated to foster the psychological health of the offspring. The skyrocketing increase in the number of children and adolescents who nowadays experience their parents’ separation poses a challenge not only to their parents and relatives but also to the family doctors, psychologists, and those required to provide specialist support in these situations.7 The evidence supported by our data linking ED to parental separation or divorce provides an additional argument for strengthening the sense of responsibility of parents and advisors before making decisions about a possible family breakdown.7 Once separation or divorce has occurred, girls in that situation can be viewed as being exposed to a higher risk of developing ED.
Lower self-esteem has been included among the disadvantages to children after parental separation,33 and a low level of self-evaluation has been also identified as a predictor of ED.2–4,16,27 Therefore, self-esteem might represent an explanation of the higher risk of ED we have found among girls whose parents were separated. Nevertheless, when we adjusted for self-esteem (and other factors) in the multivariate model, the OR decreased only from 2.16 to 1.97, meaning that other factors different from lowering self-esteem may mediate between parental disruption and the incidence of ED. This association can be explained by 2 main alternative reasons. The first is related to the construct of "life events." Life events have been previously implicated as precipitating the onset of BN. The stressful events represented by the death of 1 of both parents (widowing) or by the change in the family structure (because a parent left the home or a new stepparent arrived) were in fact defined as events considered as exposures in a previous matched case-control study that found significant associations.25
The second reason is that single-parent families may be more likely to lack the ability to educate girls regarding food habits and eating patterns, precisely because of the absence of 1 of the 2 parents during meals. This parental input is believed to exert a very important role in preventing eating disorders.8,9
Sociocultural factors are very likely to be responsible for the increasing rates of ED among adolescent girls. In fact, an environmental model to address the primary prevention of ED was proposed recently.1 Among factors in the cultural environment, the influence of mass media usually transmitting unrealistically thin ideals have been most frequently implicated.11,12,34 Nevertheless, there is little research directly assessing the association between the frequency or time spent reading, listening to, or watching mass media and the future incidence of ED. The exposure assessment in the only prospective study that we are aware of12 was based on questions related to the self-perception of the influences of media (trying to look as actresses or models in movies, television, or magazines) and did not report any association with the time spent in these activities. As self-figure perception (and therefore comparisons with the body shapes of others) may be distorted as an early consequence of ED, we believe that it is more appropriate simply to assess as an exposure variable the time spent receiving the influence of mass media and not to ask questions regarding the self-perception of the media influences.
After adjusting for confounding, we did not find any significant association with the time spent watching television. In agreement with our results, a previous cross-sectional study found that the amount of television watched was not associated with body dissatisfaction or drive for thinness, but category of program did.34 However, as recently reported,35 television viewing is a cause of increased body fatness, and we have also found in our data an association (albeit weak) between BMI and hours of television watched both during a typical weekday and during weekends (correlation coefficients: r = 0.06, P = .02 and r = 0.05, P = .019, respectively). It is worth noting that any dieting (whether the person is normal weight or obese) without proper medical follow-up may lead to an ED.
We found an association with more frequently reading printed media (specifically teen girls’ magazines) and/or more time spent listening to the radio. These associations underline the need to do an overall assessment of the influence of mass media in this kind of research. A heavier use of the printed media has been previously found to be associated with a higher propensity to weight concerns in a cross-sectional study.24 Likewise, music videos (but not amount of television watching) predicted drive for thinness in another cross-sectional study.34 Therefore, there is ground for our data to support the influence of mass media on a higher risk of developing an ED. The likely explanation is that these media are transmitting to young girls an exaggerated pressure to be thin. An obvious consequence is the recommendation that the messages transmitted by images of actresses and models in printed media must be subjected to realistic standards of body size. Moreover, an in-depth analysis of the overall culture transmitted by radio programs and magazines for young girls is required to avoid the ideal of "perfect body shape." These media should convey other messages about lifestyles and human values that are as least as important as body size.
Some limitations of our design and methods should be acknowledged. Our follow-up was not very long (18 months). This shortcoming represents an important reason to temper the potential causal interpretation of our findings because some of the "new" cases of ED that occurred during the follow-up may in fact represent subclinical preexisting cases that were false negatives in the first assessment. Additional studies with a longer follow-up are desirable to confirm our findings.
In addition, the proportion of girls who were followed up successfully (92%) deserves consideration because of the losses to follow-up. However, a higher loss to follow-up is more likely to happen when a representative sample is selected to form a cohort instead of selecting a convenience sample. Furthermore, this attrition rate (8% of ED-free girls at baseline who were lost to follow-up) is not likely to cause a bias of large magnitude. Twelve of the 90 incident cases were not diagnosed by psychiatrist-conducted interviews, but their diagnoses were based on the answers to the EAT and other questionnaires. Nonetheless, when we excluded these cases from our statistical analyses, the point estimates of the ORs did not substantially change. We did not separately assess the risk factors for AN, BN, and EDNOS, because the study did not have enough statistical power to conduct separate analyses. Future, larger, prospective studies should differentiate the risk factors for each entity.
|
ACKNOWLEDGMENTS |
|---|
This work was supported in part by grants from Fundación Echebano, the Department of Health (Navarra Regional Government Project 24/99), and Banco Santander-Central-Hispano.
|
FOOTNOTES |
|---|
Received for publication Apr 11, 2002; Accepted Jul 31, 2002.
Reprint requests to (M.A.M-G.) Unidad de Epidemiología y Salud Pública, Facultad de Medicina, Universidad de Navarra, Irunlarrea 1, E-31080 Pamplona, Spain. Email: mamartinez@unav.es
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Austin SB. Prevention research in eating disorders: theory and new directions. Psychol Med.2000; 30 :1249 –1262[CrossRef][Web of Science][Medline]
- Fairburn CG, Welch SL, Doll HA, Davies BA, O’Connor ME. Risk factors for bulimia nervosa. A community-based case-control study.
Arch Gen Psychiatry.1997; 54
:509
–517
[Abstract/Free Full Text] - Fairburn CG, Doll HA, Welch SL, Hay PJ, Davies BA, O’Connor ME. Risk factors for binge eating disorder: a community-based, case-control study.
Arch Gen Psychiatry.1998; 55
:425
–432
[Abstract/Free Full Text] - Fairburn CG, Cooper Z, Doll HA, Welch SL. Risk factors for anorexia nervosa: three integrated case-control comparisons.
Arch Gen Psychiatry.1999; 56
:468
–476
[Abstract/Free Full Text] - Patton GC, Selzer R, Coffey C, Carlin JB, Wolfe R. Onset of adolescent eating disorders: population based cohort study over 3 years.
BMJ.1999; 318
:765
–768
[Abstract/Free Full Text] - Stettler N, Tershakovec AM, Leonard MB. Onset of adolescent eating disorders. Dieting may be an early sign, rather than a cause, of eating disorder.
BMJ.1999; 318
:1761
–1762
[Free Full Text] - Tripp JH, Cockett M. Parents, parenting, and family breakdown.
Arch Dis Child.1998; 78
:104
–108
[Free Full Text] - Smolak L, Levine MP, Schermer F. Parental input and weight concerns among elementary school children. Int J Eat Disord.1999; 25 :263 –271[CrossRef][Web of Science][Medline]
- Wichstrom L. Social, psychological and physical correlates of eating problems. A study of the general adolescent population in Norway. Psychol Med.1995; 25 :567 –579[Web of Science][Medline]
- Borzekowski DL, Robinson TN, Killen JD. Does the camera add 10 pounds? Media use, perceived importance of appearance, and weight concerns among teenage girls. J Adolesc Health.2000; 26 :36 –41[CrossRef][Web of Science][Medline]
- Cusumano DL, Thompson JK. Media influence and body image in 8–11-year-old boys and girls: a preliminary report on the multidimensional media influence scale. Int J Eat Disord.2001; 29 :37 –44[CrossRef][Web of Science][Medline]
- Field AE, Camargo C, Taylor CB, Berkey CS, Colditz GA. Relation of peer and media influences to the development of purging behaviors among preadolescent and adolescent girls.
Arch Pediatr Adolesc Med.1999; 153
:1184
–1189
[Abstract/Free Full Text] - Hall A, Hay PJ. Eating disorder patient referrals from a population region 1977–1986. Psychol Med.1991; 21 :697 –701[Web of Science][Medline]
- Lucas AR, Crowson CS, O’ Fallon M, Melton J. The ups and downs of anorexia nervosa. Int J Eat Disord.1999; 26 :397 –405[CrossRef][Web of Science][Medline]
- Pérez-Gaspar M, Gual P, de Irala-Estévez J, Martínez-González MA, Lahortiga F, Cervera S. Prevalencia de Trastornos de la Conducta Alimentaria (TCA) en las adolescentes navarras (Spanish) [Prevalence of eating disorders in a representative sample of female adolescents from Navarra]. Med Clin (Barc).2000; 114 :481 –486
- Gual P, Pérez-Gaspar M, Martínez-González MA, Lahortiga F, De Irala-Estévez J, Cervera S. Self-esteem, personality, and eating disorders: baseline assessment of a prospective population-based cohort. Int J Eat Disord.2002; 31 :261 –273[CrossRef][Web of Science][Medline]
- Garner DM, Garfinkel PE. The Eating Attitudes Test: an index of the symptoms of anorexia nervosa. Psychol Med.1979; 9 :273 –279[Web of Science][Medline]
- Castro J, Toro J, Salamero M, Guimerá E. The Eating Attitudes Test. Validation of the Spanish version. Psychol Assess.1991; 7 :175 –190
- Musitu G, Garcia F, Gutiérrez M. Manual del AFA. Autoconcepto Forma-A. (Spanish) [Manual of Self-Concept Form A]. Madrid, Spain: TEA Ediciones; 1994
- Garner DM, Olmsted MP, Polivy J. Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. Int J Eat Disord.1983; 2 :15 –34
- Hosmer DW, Lemeshow SA. Applied Logistic Regression. 2nd ed. New York, NY: Wiley; 2000
- Zipfel S, Lowe B, Reas DL, Deter HC, Herzog W. Long-term prognosis in anorexia nervosa: lessons from a 21-year follow-up study. Lancet.2000; 355 :721 –722[CrossRef][Web of Science][Medline]
- Ben-Tovim DI, Walker K, Gilchrist P, Freeman R, Kalucy R, Esterman A. Outcome in patients with eating disorders: a 5-year study. Lancet.2001; 357 :1254 –1257[CrossRef][Web of Science][Medline]
- Field AE, Cheung L, Wolf AM, Herzog DB, Gortmaker SL, Colditz GA. Exposure to the mass media and weight concerns among girls. Pediatrics.1999; 103(3) . Available at: www.pediatrics.org/cgi/content/full/103/3/e36
- Welch, SL, Doll HA, Fairburn CG. Life events and the onset of bulimia nervosa: a controlled study. Psychol Med.1997; 27 :515 –522[CrossRef][Web of Science][Medline]
- Shisslak CM, Crago M, McKnight KM, Estes LS, Gray N, Parnaby OG. Potential risk factors associated with weight control behaviors in elementary and middle school girls. J Psychosom Res.1998; 44 :301 –313[CrossRef][Web of Science][Medline]
- Button EJ, Sonuga-Barke EJS, Davies J, Thompson M. A prospective study of self-esteem in the prediction of eating problems in adolescent schoolgirls: questionnaire findings. Br J Clin Psychol.1996; 35 :193 –203
- Killen JD, Taylor CB, Hayward C, et al. Weight concerns influence the development of eating disorders: a 4-year prospective study. J Consult Clin Psychol.1996; 64 :936 –940[CrossRef][Web of Science][Medline]
- Martin GC, Wertheim EH, Prior M, Smart D, Sanson A, Oberklaid F. A longitudinal study of the role of childhood temperament in the later development of eating concerns. Int J Eat Disord.2000; 27 :150 –162[CrossRef][Web of Science][Medline]
- Santonastaso P, Friederici S, Favaro A. Full and partial syndromes in eating disorders: a 1-year prospective study of risk factors among female students. Psychopathology.1999; 32 :50 –56[CrossRef][Web of Science][Medline]
- Hsu LK. Can dieting cause an eating disorder? Psychol Med.1997; 27 :509 –513[CrossRef][Web of Science][Medline]
- Shisslak CM, Crago M, Estes LS. The spectrum of eating disturbances. Int J Eat Disord.1995; 18 :209 –219[Web of Science][Medline]
- Ferri E, ed. Life at 33: The Fifth Follow-up of the National Child Development Study. London, United Kingdom: National Children’s Bureau and City University Press; 1993
- Tiggemann M, Pickering AS. Role of television in adolescent women’s body dissatisfaction and drive for thinness. Int J Eat Disord.1996; 20 :199 –203[CrossRef][Web of Science][Medline]
- Robinson TN. Television viewing and childhood obesity. Pediatr Clin North Am.2001; 48 :1017 –1025[CrossRef][Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2003 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  A. E. Field, K. M. Javaras, P. Aneja, N. Kitos, C. A. Camargo Jr, C. B. Taylor, and N. M. Laird Family, Peer, and Media Predictors of Becoming Eating Disordered Arch Pediatr Adolesc Med, June 1, 2008; 162(6): 574 - 579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R F Drewett The social facilitation of food intake Arch. Dis. Child., May 1, 2007; 92(5): 377 - 377. [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Wilson, R. Peebles, K. K. Hardy, and I. F. Litt Surfing for Thinness: A Pilot Study of Pro-Eating Disorder Web Site Usage in Adolescents With Eating Disorders Pediatrics, December 1, 2006; 118(6): e1635 - e1643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Field, S. B. Austin, C. A. Camargo Jr, C. B. Taylor, R. H. Striegel-Moore, K. J. Loud, and G. A. Colditz Exposure to the Mass Media, Body Shape Concerns, and Use of Supplements to Improve Weight and Shape Among Male and Female Adolescents Pediatrics, August 1, 2005; 116(2): e214 - e220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Joffe When Perception Is Reality Arch Pediatr Adolesc Med, June 1, 2005; 159(6): 592 - 593. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Newton Eating alone, parents' marital status, and use of radio and girls' magazines were risk factors for eating disorders Evid. Based Nurs., October 1, 2003; 6(4): 120 - 120. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
