 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, Minnesota
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PARTICIPANTS AND METHODS. Data are from Project EAT (Eating Among Teens), a 5-year longitudinal study of eating, activity, weight, and related variables in 2516 middle and high school students. In 1999 (time 1), participants completed surveys and had their height and weight measured. In 2004 (time 2), participants were resurveyed.
RESULTS. For female adolescents, the frequency of healthy, unhealthy, and extreme weight-control behaviors increased with increasing magazine reading after adjusting for time 1 weight-control behaviors, weight importance, BMI, and demographic covariates. The odds of engaging in unhealthy weight-control behaviors (such as fasting, skipping meals, and smoking more cigarettes) were twice as high for the most frequent readers compared with those who did not read magazine articles about dieting and weight loss. The odds of using extreme weight-control behaviors (such as vomiting or using laxatives) were 3 times higher in the highest frequency readers compared with those who did not read such magazines. There were no significant associations for either weight-control behaviors or psychological outcomes for male adolescents.
CONCLUSIONS. Frequent reading of magazine articles about dieting/weight loss strongly predicted unhealthy weight-control behaviors in adolescent girls, but not boys, 5 years later. Findings from this study, in conjunction with findings from previous studies, suggest a need for interventions aimed at reducing exposure to, and the importance placed on, media messages regarding dieting and weight loss.
Key Words: mass media • diets • weight control • mental health • adolescents • longitudinal studies
Abbreviations: EAT—Eating Among Teens • SES—socioeconomic status • OR—odds ratio • CI—confidence interval
Unhealthy weight-control practices, clinical eating disorders, and subclinical disordered eating are significant problems for adolescents, especially adolescent girls. In the 2003 Youth Risk Behavior Survey,1 for example, 
1 in 10 of the 15240 high school students surveyed reported having taken diet pills, powders, or liquids in the previous 30 days (11% of girls and 7% of boys). Thirteen percent of the students (9% of boys and 14% of girls) reported fasting for 
24 hours to lose or maintain weight, and 6% reported vomiting or taking laxatives to lose or maintain weight (8% of girls and 4% of boys). Likewise, in Project Eating Among Teens (EAT), a study of 4746 middle and high school students, 12% of adolescent girls reported vomiting or taking diet pills, laxatives, or diuretics to lose weight.2 These types of unhealthy weight-control practices have been found to be associated with adverse psychological outcomes,3,4 physical outcomes, including weight gain over time, and nutritional outcomes, such as inadequate dietary intake.5–7 Unhealthy weight-control behaviors have been found to predict more severe disordered eating in adolescents as well.5 Prevalence rates for bulimia and anorexia are 1% to 2% of female adolescents and adults, and partial syndrome rates as high as 15% have been reported.8 Eating disorders can have significant medical complications, such as heart problems, electrolyte imbalances, and even death, with anorexia having a mortality rate of 5.6% per decade.9
Researchers have identified multiple concurrent and prospective predictors of problematic eating behavior, including BMI, body dissatisfaction, and perceived pressure to be thin from peers, family, and the media.10 Studies of media exposure, the topic of the current investigation, have had inconsistent findings. For instance, some cross-sectional studies have found that adolescents' self-reports of their consumption of media that promote the thin ideal were correlated with eating disorder symptoms,11,12 whereas others have not found a significant association13 or have had mixed results.14 One prospective study of thin ideal-promoting media use in young adolescent girls found that decreases in magazine reading over 16 months were associated with decreases in eating disordered symptoms,15 suggesting that media use is related to unhealthy eating behaviors. Furthermore, a meta-analysis of randomized experiments indicated that exposure to media images that depict extremely thin women leads to increased body dissatisfaction,16 which is an established risk factor for eating disorders.10 However, these randomized experiments assessed only short-term elevations in problematic outcomes. Furthermore, studies to date have primarily focused on the effects of exposure to thin-ideal images, whereas less research has examined the influence on adolescents of media information about dieting and weight loss. In addition, very few studies have examined this issue in male adolescents.
An earlier investigation that examined frequency of reading dieting/weight loss magazine articles cross-sectionally in the first wave of Project EAT data found that such magazine reading was strongly associated with healthy and unhealthy weight-control behaviors, as well as psychological variables (self-esteem, depressive symptoms, and body image), in both boys and girls.17 However, given the cross-sectional nature of the Utter et al17 study, conclusions regarding temporality could not be drawn. It is possible that adolescents who are interested in weight control and who are dieting would seek out information to help them in this pursuit, perhaps in the form of magazine articles about dieting and weight loss, leading to an association between magazine reading and weight-control behaviors. Thus, a longitudinal study is needed to determine whether adolescents' choice of reading materials precedes unhealthy weight-control behaviors.
This study builds on previous research by examining whether reading magazine articles about dieting or weight loss predicts eating behaviors and psychological outcomes over 5 years during adolescence. We hypothesized that adolescents' self-report of frequency of reading magazine articles about weight loss or dieting at time 1 would significantly predict time 2 healthy and unhealthy weight-control behaviors, binging, body satisfaction, depressive symptoms, and self-esteem, after adjustment for the time 1 level of the behavior and demographic covariates. To address investment in weight and appearance as a possible confounder that might drive both magazine article reading and weight-control behaviors and psychological outcomes, we also adjusted for baseline importance of weight to the participant.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sample
Participants were recruited from 31 Minnesota public junior and senior high schools, corresponding to a younger cohort (from junior high schools) and an older cohort (from high schools). The study sample at time 1 consisted of 4746 adolescents in grades 7 through 12, with approximately equal numbers of boys and girls. Project EAT-II obtained time 2 data during the 2003–2004 school year, enrolling 2516 adolescents from the original sample. Of the original sample at time 1, 1074 (22.6%) were lost to follow-up for various reasons, primarily that they were missing contact information at time 1 or that the address they gave was not valid at time 2. Of the remaining 3672 participants with valid addresses to whom surveys were mailed, 2516 responded, representing 53.0% of the original cohort and 68.4% of participants who could be contacted for Project EAT-II. The final study population consisted of 1130 male participants (45%) and 1386 female participants (55%) who completed surveys at both time 1 and time 2. One third of the participants (32%) were in the younger cohort; at time 1 their mean age was 12.8 years (SD: 0.8), and at time 2 their mean age was 17.2 years (SD: 0.6). Two thirds (68%) of the participants were in the older cohort; at time 1 their mean age was 15.8 years (SD: 0.8), and at time 2 their mean age was 20.4 years (SD: 0.8). The weighted ethnic/racial and socioeconomic status (SES) proportions are as follows: 48.3% white, 18.9% black, 5.8% Hispanic, 19.6% Asian, 3.6% Native American, and 3.8% mixed or other race, whereas SES was low (17.8%), middle-low (18.9%), middle (26.7%), middle-high (23.3%), and high (13.3%; see below for description of weighting procedures).
Measures
The entire time 1 survey contained 221 items chosen from the literature or developed for the current study by the research team in consultation with other experts in the area and with input from focus groups of adolescents.19 At time 1, pilot testing was conducted with 7th- and 10th-graders, and revisions were undertaken in compiling the final survey packet. The time 2 survey was a revised version of the time 1 survey.
Frequency of Diet/Weight Loss Magazine Article Reading
A single question, "How often do you read magazine articles in which dieting or weight loss are discussed?" was used to assess this construct. Response options included "never," "hardly ever," "sometimes," or "often." In descriptive analyses, the responses were dichotomized into low frequency of diet/weight loss magazine article reading ("never" or "hardly ever") and high frequency of diet/weight loss magazine article reading ("sometimes" or "often"). The untransformed responses were used in all of the regression analyses.
BMI and Weight Status
Anthropometric measures of height and weight were taken at time 1, using standardized equipment and procedures. Height was measured without shoes, and weight was measured in street clothes without heavy outerwear. BMI was computed using the formula: weight (in kilograms) divided by squared height (in meters). Weight status (underweight, healthy weight, at risk for overweight, and overweight) was based on Centers for Disease Control and Prevention Guidelines.20
Weight Importance
This item was adapted from Yanovski's21 Questionnaire on Eating and Weight Patterns-Revised. Participants responded to the question, "During the past 6 months, how important has your weight or shape been in how you feel about yourself?" Responses ranged from "Weight and shape were not very important" to "Weight and shape were the most important things that affected how I felt about myself."
Weight-Control Behaviors
Healthy, unhealthy, and extreme weight-control behaviors during the past year were assessed with the following question: "Have you done any of the following things to lose weight or keep from gaining weight during the past year? (yes or no for each method)." Responses categorized as healthy weight-control behaviors included the following: (1) exercised, (2) ate more fruits and vegetables, (3) ate less high-fat foods, and (4) ate less sweets. Responses categorized as unhealthy weight-control behaviors included the following: (1) fasted, (2) ate very little food, (3) used a food substitute (powder or a special drink), (4) skipped meals, and (5) smoked more cigarettes. Responses categorized as extreme weight-control behaviors included the following: (1) took diet pills, (2) made myself vomit, (3) used laxatives, and (4) used diuretics. Dichotomous indicators for each of the weight-control variables were constructed, with those reporting no relevant behaviors categorized as "none" and those reporting 1 of the relevant behaviors categorized as "any."
Binge Eating
Binge eating was assessed from a combination of 2 items, "In the past year, have you ever eaten so much food in a short period of time that you would be embarrassed if others saw you (binge-eating)?" and "During the times when you ate this way, did you feel that you couldn't stop eating or control what or how much you were eating?" Both questions were from the Questionnaire on Eating and Weight Patterns-Revised.21 Participants who answered "yes" to both questions were classified as having binged.
Body Satisfaction
Ten items from the Body Shape Satisfaction Scale22 were used to assess satisfaction with appearance. Each item consists of a rating on a 5-point scale of the level of satisfaction with 1 of 10 body parts (height, weight, body shape, waist, hips, thighs, stomach, face, body build, and shoulders). Cronbach's 
for the scale in this study was .92 at time 1 and .93 at time 2.
Depressive Symptoms
A 6-item scale of depressive symptoms developed by Kandel and Davies23 was used. This scale has demonstrated reliability and validity in previous studies with adolescent clinical and nonclinical samples.23 The internal consistency reliability in the current sample was .82 at time 1 and .81 at time 2.
Self-esteem
A shortened version of the Rosenberg Self-Esteem Scale,24 consisting of 6 items, assessed general self-esteem. This scale is a well-validated questionnaire used to assess self-esteem in adolescent and adult samples.25 A selection of items was used for the sake of brevity of the overall survey. The Cronbach's for the scale at time 1 was .79 and at time 2 was .81.
Sociodemographic Characteristics
Gender, age (in years), ethnicity/race, and SES were based on self-report at time 1. SES was calculated based on an algorithm that weighted parental education level most heavily but also took into account family eligibility for public assistance, eligibility for free or reduced-cost school meals, and employment status of the mother and father (see ref 18 for more information).
Statistical Analysis
Initial analyses included examination of the demographic characteristics of adolescents who were high-frequency readers of diet/weight loss magazine articles, using 
2 and trend analyses. Logistic regression was used to examine the association between time 1 magazine reading and time 2 healthy, unhealthy, and extremely unhealthy weight-control behaviors and binge eating. Multiple linear regression was used to evaluate the relationships between time 1 magazine reading and time 2 body satisfaction, depression, and self-esteem. The logistic and linear regression models were adjusted for demographic characteristics including race/ethnicity, cohort (younger or older), SES, BMI at time 1, age, squared age (to account for possible curvilinear relationships between age and the dependent variables), and time 1 level of the dependent variable being predicted. To address the possibility that reading magazine articles about weight loss and dieting is simply a proxy for the level of importance placed on weight, we also included time 1 weight importance as a covariate. For the logistic regressions, odds ratios (ORs) were computed comparing the "never" category with the other 3 categories of diet/weight loss magazine article reading. For the linear regressions, adjusted means were computed for each of the 4 levels of magazine reading, and the F test of the magazine variable was examined. All of the analyses were stratified by gender.
Attrition at time 2 differed across sociodemographic characteristics. Thus, in all of the analyses, participants' responses were weighted to adjust for this differential response rate. The response propensity method26 was used to generate the weights, such that an individual's propensity weight was the inverse of the estimated probability that an individual of like characteristics responded at time 2. Response propensities (ie, the probability of responding to the time 2 survey) were estimated using a logistic regression, with a large number of predictor variables from the time 1 survey predicting response at Time 2 (yes/no). The selected response propensity model included main effects for time 1 gender, native born status, ethnicity/race, SES, overweight status, parental marital status, individual's concern about health, and most common grade received in school. In addition, weights were calibrated so that the weighted total sample sizes used in analyses for each gender cohort accurately reflected the actual observed sample sizes in those groups. The weighting method resulted in estimates representative of the demographic make-up of the original time 1 sample. SAS 9.1 (SAS Institute, Inc, Cary, NC) was used for all of the analyses.27
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Weight-Control Behaviors
For female adolescents, the odds of engaging in healthy, unhealthy, and extremely unhealthy weight-control behaviors at time 2 increased with increasing frequency of time 1 magazine reading, after adjustment for age, race/ethnicity, cohort, SES, time 1 BMI, time 1 weight importance, and time 1 levels of dependent variables (Table 2). For healthy weight-control behaviors, female adolescents who reported "hardly ever" or "sometimes" reading magazines had 1.6 and 2.4 times the odds of engaging the behaviors compared with the reference group of nonreaders, but no significant increase in odds was found for female adolescents who reported "often" reading magazines. Girls who reported magazine reading had between 1.6 and 2.0 times the odds of engaging in unhealthy weight-control behaviors, compared with nonreaders of magazine articles about dieting or weight loss, in adjusted analyses (Table 2). Likewise, for extremely unhealthy weight-control behaviors, frequency of magazine reading had a strong positive association with the odds of engaging in unhealthy behaviors at time 2, with ORs ranging from 2.3 for female adolescents who reported "hardly ever" reading magazine articles to 3.2 for those who "often" read magazine articles. A marginally significant, ordered association was found between binging and magazine reading for female adolescents.
|
Psychological Variables
Table 3 contains the adjusted means and P values for the test of the relationships of time 1 frequency of diet/weight loss article reading to the time 2 continuous measures of depression, self-esteem, and body satisfaction. For both male and female adolescents, all of the relationships are nonsignificant.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The strength and persistence of magazine reading as a predictor of weight-control behaviors in female adolescents is noteworthy, particularly given the long period of time between exposure and outcome behaviors. Previous experimental research has demonstrated short-term effects of exposure to the thin ideal on body image outcomes,16 but the data regarding the longitudinal effects of media use have, to this point, been inconsistent or weak.10 Several studies13,14 have found small or weak effects of media exposure on body image and negative affect but not on eating or dieting behaviors. In contrast, Stice et al28 did find that adolescent girls who were randomly assigned to receive a subscription to a fashion magazine had increased negative affect, dieting, bulimic symptoms, and body dissatisfaction, although this effect was found only in more vulnerable girls who had lower levels of social support or previously higher levels of body dissatisfaction or perceived pressure to be thin.
Regarding previous research with this sample, our findings are in line with the cross-sectional findings from Project EAT for female adolescents but not for male adolescents.17 Although cross-sectionally, adolescent boys' reading of dieting or weight loss articles was related to weight-control behaviors, we did not find this relationship longitudinally. The associations were generally stronger for female than for male adolescents cross-sectionally, and it is possible that the weaker effect among male adolescents was attenuated over time. Alternatively, the question about dieting and weight loss articles may not have assessed the muscular ideal for male adolescents as well as it did the thin ideal for female adolescents.29 In addition, although cross-sectionally there was a relationship between magazine reading and psychological indicators in both male and female participants,17 longitudinally no association was found in either gender. This difference suggests that reading articles about dieting and weight loss is a specific risk factor for problematic eating patterns, including extreme behaviors, such as vomiting and laxative use, but is not a nonspecific predictor of general distress. It is somewhat surprising that associations were found with weight-control behaviors but not with body satisfaction. This may be because of the fact that our magazine reading item asked specifically about articles on dieting and weight loss.
Strengths of the current investigation include its large, diverse sample and its longitudinal approach. Using longitudinal data, we were able to examine the temporality of the relationship between media use and eating and other problems. The follow-up was over a long period of time and captured developmental periods that are important for the development and maintenance of eating and weight-related behaviors.30 Limitations of the study include the use of a single item to assess magazine article reading, providing no information regarding the types of magazines read or frequency of magazine reading during the 5-year study period. A more comprehensive assessment of media use at multiple time points, including broadcast media and print media, would add to our understanding of the relationship between media use and eating and weight-related behaviors and psychological outcomes.
In addition, although the significant effects seen here were constrained to adolescent girls, the possible relationships in adolescent boys have yet to be fully explored. Researchers are beginning to discover that the predictors of body-enhancing behaviors among male adolescents may be different than that in female adolescents.29 Furthermore, measures of body image originally developed for female adolescents may not capture the construct as precisely in male adolescents, because body image in male adolescents seems to include both high muscularity and low body fat factors, whereas for female adolescents the focus is on thinness.29 To fully explore the associations between media and body image in male adolescents, future research should incorporate these findings into the assessment of body image in male adolescents.
The results of this study should be considered in combination with other relevant studies. A large experimental literature has found robust effects of exposure to thin-ideal images in the media on body dissatisfaction and negative affect,16 and the experimental study of Stice et al28 showed increases in negative affect, dieting, bulimic symptoms, and body dissatisfaction in vulnerable adolescent girls randomly assigned to receive a fashion magazine subscription. The results of the current study, when combined with this literature, highlight the importance of media in influencing adolescent girls' physical and emotional health and suggest a need for interventions to moderate this effect. Possible avenues for intervention might include decreasing female adolescents' exposure to potentially harmful media messages and enhancing media literacy skills. It might be helpful, for instance, to discuss with parents limiting their daughter's access to magazines that heavily portray the thin ideal and promote dieting. A first step in this direction might consist of recommending that mothers stop buying these magazines for themselves so that they are not inadvertently exposing their daughters to the dieting/weight loss messages and thin ideal images in these magazines. The impact of this type of intervention on media use and, ultimately, on unhealthy eating attitudes and behaviors, would need to be assessed. Another approach to this problem involves enhancing media literacy skills. Several studies have shown existing media literacy interventions to be effective in reducing adolescents' vulnerability to thinness promoting media messages (see ref 31 for a review of these interventions). Finally, implications for health care providers include the advisability of removing appearance, diet, and weight loss-oriented magazines from waiting areas and examination rooms, so as not to reinforce potentially harmful social norms and messages within health promotion settings.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Address correspondence to Patricia van den Berg, PhD, Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, 1300 South Second St, Suite 300, Minneapolis, MN 55454. E-mail: vandenberg{at}epi.umn.edu
The author have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. J. Howe, A. F. Jawad, S. D. Kelly, and T. H. Lipman Weight-Related Concerns and Behaviors in Children and Adolescents With Type 1 Diabetes Journal of the American Psychiatric Nurses Association, January 1, 2008; 13(6): 376 - 385. [Abstract] [PDF]
|
|
|
|
 |
|
 Dieting Advice: What's Your Waiting Room Saying? Journal Watch Pediatrics and Adolescent Medicine, February 28, 2007; 2007(228): 4 - 4. [Full Text]
|
|
Read all P3Rs
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2007 American Academy of Pediatrics. All rights reserved. |
||
|
||
