- Email: [email protected]
Weight-control behaviors among adolescent girls and boys: implications for dietary intake
RESEARCH Current Research
Weight-Control Behaviors among Adolescent Girls and Boys: Implications for Dietary Intake DIANNE NEUMARK-SZTAINER, PhD, MPH, RD; PETER J. HANNAN, MStat; MARY STORY, PhD, RD; CHERYL L. PERRY, PhD
ABSTRACT Objective To examine associations between healthful and unhealthful weight-control behaviors and dietary intake among adolescents. Design Cross-sectional survey (Project EAT [Eating Among Teens]). Subjects/Setting The study population included 4,144 middle and high school students from Minneapolis/St. Paul public schools from diverse racial and socioeconomic backgrounds. Statistical Analyses Dietary intake patterns were compared across adolescent girls and boys reporting unhealthful, only healthful, or no weight-control behaviors in unadjusted analyses and analyses adjusted for sociodemographic factors and energy intake. Results Among girls, mean intakes differed across weightcontrol behaviors for all foods and nutrients examined. P values ranged from P⫽.006 to P⬍.001. Girls using unhealthful weight-control behaviors had significantly lower intakes of fruit; vegetables; grains; calcium; iron; vitamins A, C, and B-6; folate; and zinc than girls using only healthful weight-control behaviors. Compared with girls reporting no weight-control behaviors, girls using unhealthful weight-control behaviors had lower intakes of grains, calcium, iron, vitamin B-6, folate, and zinc. In contrast to the girls, boys reporting unhealthful weightcontrol behaviors did not have poorer dietary intakes than boys not using weight-control behaviors or using only healthful behaviors. Among boys, there were no significant differences in mean intakes of vegetables; grains; calcium; iron; vitamins A, C, and B-6; folate; and zinc. Furthermore, boys using unhealthful weight-control behaviors had higher fruit intakes (P⫽.002) than boys reporting no weight-control behaviors. Conclusions Adolescent girls who engage in unhealthful weight-control behaviors are at increased risk for dietary inadequacy. The findings demonstrate a need for interventions to prevent unhealthful weight-control behaviors D. Neumark-Sztainer, P. J. Hannan, M. Story, and C. L. Perry are with the Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis. Address correspondence to: Dianne Neumark-Sztainer, PhD, MPH, RD, Division of Epidemiology, School of Public Health, University of Minnesota, 1300 S Second St, Suite 300, Minneapolis, MN 55454. E-mail: Neumark@ epi.umn.edu Copyright © 2004 by the American Dietetic Association. 0002-8223/04/10406-0006$30.00/0 doi: 10.1016/j.jada.2004.03.021
© 2004 by the American Dietetic Association
in adolescent girls, and to promote healthful weight-control behaviors when indicated. J Am Diet Assoc. 2004;104:913-920.
D
ata from numerous studies have demonstrated a high prevalence of weight-control behaviors among adolescents, particularly adolescent girls (1-3). In previously reported findings from the Project EAT (Eating Among Teens) study, 45% of adolescent girls and 21% of adolescent boys reported that they were currently trying to lose weight, and an additional 26% of girls and 23% of boys were trying to maintain their weight (4,5). Of particular concern were the high percentages of youth who reported using unhealthful weight-control behaviors. Unhealthful weight-control behaviors such as skipping meals, fasting, using food substitutes, or smoking more cigarettes were reported by more than half (57%) of the girls and by one third (33%) of the boys. An additional 12% of girls and 5% of boys reported extremely unhealthful behaviors such as use of diet pills, laxatives, diuretics, or vomiting. Some weight-control behaviors (eg, increasing fruits and vegetables and decreasing foods high in fats and sugars) are appropriate, even desirable, for most adolescents, when done in moderation. These behaviors have the potential to help adolescents achieve a healthful body weight through the prevention of excess weight gain or, when appropriate, through gradual weight loss. However, because of the importance of adequate nutrition during adolescence (6-9), it is essential to determine the potential implications of different weight-control behaviors on dietary intake in youth. On one hand, there may be positive implications if adolescents become more attentive eaters, consume more fruits, vegetables, and whole grains, and avoid excessive fat intake. On the other hand, there may be negative implications if energy restrictions or selective food choices lead to decreased intake of nutrients such as iron and calcium, which are important during this period of growth and development (10,11). To guide the development of appropriate messages and dietary intervention programs for youth, it is important to determine whether weight-control behaviors are associated with dietary intake and whether associations vary for different types of weight-control behaviors. In general, studies have suggested that youth engaging in weight-control behaviors (unhealthful weight-control behaviors in particular), are at risk for inadequate dietary intakes (12-17), although in some studies associations were not found (18-20), and in other studies weightcontrol behaviors were associated with more positive eating patterns such as decreased consumption of high-
Journal of THE AMERICAN DIETETIC ASSOCIATION
913
fat baked goods (ie, cookies, doughnuts, pie, or cake) (21), sweets, salty snacks, and soft drinks (22). Inconsistencies may be due to differences in measures used to assess weight-control behaviors, dietary assessment tools, dietary components being examined, and sample characteristics and sizes. Many of the studies that have examined associations between weight-control behaviors and dietary intake did not distinguish between different types of weight-control behaviors (16,18,21,23), used inadequate measures of dietary intake (15,16,21-23), had small samples (or small numbers of dieters within samples) (12-14,19), or included only girls in their study populations (12,17,20). The current study explores associations between weight-control behaviors and dietary intake. It expands on existing studies exploring these associations in that data are drawn from a large, ethnically diverse population of adolescents. In addition, healthful weight-control behaviors, unhealthful weight-control behaviors, and a broad array of dietary outcomes are examined. Specifically, the current study compares intake patterns of key foods and nutrients among adolescent girls and boys reporting no weight-control behaviors in the past year, only healthful weight-related behaviors, or unhealthful weight-control behaviors. Mean dietary intakes and percentages of adolescent girls and boys meeting dietary recommendations are examined across these three groups. The current study further explores whether associations between weight-control behaviors and dietary intake differ across race/ethnicity and weight status of adolescent girls and boys. METHODS Study Population and Study Design The overall study population included 4,746 adolescents from 31 public middle schools and high schools from urban and suburban school districts in the St Paul/Minneapolis area of Minnesota who participated in Project EAT. Trained research staff administered surveys within school classes and assessed height and weight within a private area. Study procedures were approved by the University of Minnesota Human Subjects Committee and by the research boards of the participating school districts. Consent procedures were done in accordance with the requests of the participating school districts; in some schools passive consent procedures were utilized, whereas active consent procedures were required in others. The response rate for student participation was 81.5%. Further details on the study sample and design have been previously published (24-26). Participants who did not complete the food frequency questionnaire (n⫽344) were not included in the current analysis. In addition, participants who had energy intakes less than 400 kcal/day or more than 7,000 kcal/day (n⫽258) were excluded from analyses examining associations with dietary intake because these values are considered biologically implausible for habitual intake. Thus, the final study population in the current analysis included 4,144 participants. The adolescents included in the current analysis were compared with those who were excluded due to missing or unusable dietary intake data and found to be similar in
914
June 2004 Volume 104 Number 6
terms of sex and body mass index (BMI), but more likely to be white, of higher socioeconomic status, and in middle school (rather than high school). These differences were not expected to strongly influence associations between weight control practices and dietary intake that are being explored in the analyses. Measures Weight-control behaviors over the past year were assessed with the question: “Have you done any of the following things in order to lose weight or keep from gaining weight during the past year?” (yes/no for each method). Responses classified as healthful weight-control behaviors included lifestyle behaviors recommended for all adolescents for overall health promotion and weight management: (a) exercise, (b) ate more fruits and vegetables, (c) ate less high-fat foods, and (d) ate less sweets. Responses classified as unhealthful weight-control behaviors included: (a) fasted, (b) ate very little food, (c) used food substitute (powder/special drink), (d) skipped meals, (e) smoked more cigarettes, (f) took diet pills, (g) made myself vomit, (h) used laxatives, and (i) used diuretics. Respondents were classified as using healthful weight-control behaviors if they reported any of the healthful behaviors and no unhealthful behaviors. They were categorized as using unhealthful weight-control behaviors if they reported any of the unhealthful behaviors. Most of the youth using unhealthful weight-control behaviors also reported the use of healthful weight-control behaviors; therefore, mutually exclusive groups were not possible. For the majority of specific weight-control behaviors used in the past year, test-retest values over a 2-week interval ranged from 0.50 to 0.68, but lower values were found for laxatives (0.29) and food substitutes (0.44). Dietary intake was assessed with the Youth and Adolescent Food Frequency Questionnaire (27,28). Servings of fruits, vegetables (including and not including french fries), grains (eg, cereal, bread, pasta, rice), and sweet and salty snack foods (eg, potato chips, candy, cake, and cookies) were examined in addition to the following nutrients: energy (kcal), total fat (% of total energy), saturated fat (% of total energy), protein (% of total energy), carbohydrate (% of total energy), calcium (mg), iron (mg), vitamin A (IU), vitamin C (mg), vitamin B-6 (mg), folate (g), and zinc (mg). In addition to examining mean intakes, we also examined the percentages of youth meeting the Healthy People 2010 objectives for fruits, vegetables, and grains (29), and the Dietary Reference Intakes (DRI) for macronutrients and micronutrients (30-32). Sex, school level, ethnicity/race, and socioeconomic status (SES) were based on self-report (24). BMI was based on height and weight measurements taken by research staff using standardized equipment and procedures. Data Analysis Girls and boys were analyzed separately due to sex differences in weight-control behaviors. Analyses of variance generated mean outcomes for adolescents reporting no weight-control behaviors, only healthful weight-control behaviors, or unhealthful weight-control behaviors.
To allow for the clustered design, the school was included in all analyses as a random effect. For skewed outcomes such as servings of fruits and vegetables, the square root of the outcome was analyzed; however, means are reported on the natural scale for all outcomes. The percent of students meeting the various Year 2010 objectives were treated as continuous measures. Analysis was by PROC MIXED (SAS, Release 8.2). Covariates used in adjusted analyses were grade (middle vs high school), ethnicity/race (white, African American, Hispanic, Asian American, other), and SES (five levels). Analyses examining mean intakes of foods and nutrients were further adjusted for energy intake, but analyses examining the percentages of adolescents eating according to dietary recommendations were not energyadjusted because most dietary recommendations are consistent across energy intake. Further energy adjustments for macronutrients were not done because they are reported as percentages of total energy intake in all analyses. Results from unadjusted analyses are presented in the tables; findings from adjusted analyses are discussed in the text. The possibility of confounding of the relationship between dietary intake and weight control methods by race/ ethnicity was examined by including the interaction of race/ethnicity (five categories) with weight-control behavior (three categories) in the adjusted analyses. A significance level of less than .01 was required for the global F-test [8 degrees of freedom (df)]. Where the global test was significant, more specific post hoc hypotheses were examined comparing each of the four minority racial/ ethnic groups against the majority white group (2 df). Adjusted means were generated to describe the pattern of interaction. For weight status as a possible confounder, overweight was defined as BMI at or more than the 85th percentile, and the main effect for overweight and interaction of overweight with weight-control methods were added to the adjusted model. Again, a significance level less than .01 was required for the global test of interaction (2 df test). RESULTS Description of Study Population Table 1 provides a description of the study population in terms of age, BMI, ethnicity/race, SES, and weight-control behaviors. More than half of the girls (57.2%) and nearly one third of the boys (31.6%) reported the use of at least one unhealthful weight-control behavior over the past year. Mean Dietary Intake by Weight-Control Behaviors In general, girls using unhealthful weight-control behaviors had poorer dietary intakes than girls reporting no weight-control behaviors or only healthful behaviors (Table 2). Girls using unhealthful weight-control behaviors had significantly lower intakes of fruit, vegetables, grains, and all micronutrients examined (calcium; iron; vitamins A, C, and B-6; folate; and zinc) than girls using only healthful weight-control behaviors. Compared with girls reporting no weight-control behaviors, girls using unhealthful weight-control behaviors had lower intakes
Table 1. Description of study population: Body mass index, sociodemographic characteristics, and weight-control behaviors Adolescent girls (Nⴝ2,064a) Age (y) BMIc Race/ethnicity White African American Hispanic Asian American Native American Mixed/other SESd High High-middle Middle Low-middle Low Grade level Middle school High school Weight-control behaviors None Healthful only Unhealthful
Adolescent boys (Nⴝ2,080a)
Mean 14.8 23.2
SDb 1.67 4.94
N
%
N
%
997 340 100 426 85 86
49.1 16.7 4.9 20.9 4.2 4.2
1,120 316 123 377 58 61
54.4 15.4 6.0 18.4 2.8 3.0
274 448 519 382 386
13.7 22.3 25.8 19.0 19.2
297 528 552 372 278
14.7 26.0 27.2 18.4 13.7
680 1,358
33.4 66.6
675 1,385
32.8 67.2
253 622 1,168
12.4 30.5 57.1
610 798 650
29.6 38.8 31.6
Mean 14.9 23.0
SD 1.65 4.81
a Due to missing values for certain variables, numbers do not always add up to this maximum N. b SD⫽standard deviation. c BMI⫽body mass index. d SES⫽socioeconomic status.
of grains and certain micronutrients (calcium, iron, vitamin B-6, folate, and zinc). Analyses adjusting for ethnicity/race, SES, and school level found similar patterns (data not shown in tables). In energy-adjusted analyses, differences between girls using unhealthful and healthful weight-control behaviors remained statistically significant for calcium, iron, vitamins A and B-6, folate, and zinc, with lower intakes among girls using unhealthful behaviors, but differences between girls using unhealthful behaviors and no weightcontrol behaviors were no longer statistically significant for these nutrients (data not shown). Girls using only healthful weight-control behaviors had higher fruit intake than girls not using weight-control behaviors and higher vitamin A intake than both of the other groups. Associations between weight-control behaviors and servings of vegetables, grains, snack foods, and vitamin C were no longer statistically significant. Other findings from energy-adjusted analyses were similar to those shown in Table 2 for unadjusted analyses. In contrast to the girls, boys reporting unhealthful weight-control behaviors did not have poorer dietary in-
Journal of THE AMERICAN DIETETIC ASSOCIATION
915
Table 2. Dietary intake (mean/day) by types of weight-control behaviors used in the past year in adolescent girls Weight-Control Behaviors None (Nⴝ253) Energy (kcal) Fruits (servings) Vegetables (servings not including french fries) Vegetables (servings including french fries) Grains (servings) Sweet/salty snack foods (servings) Total fat (% of energy) Saturated fat (% of energy) Protein (% of energy) Carbohydrate (% of energy) Calcium (mg) Iron (mg) Vitamin A (IU) Vitamin C (mg) Vitamin B-6 (mg) Folate (g) Zinc (mg)
Healthful only (Nⴝ622)
Unhealthful (Nⴝ1,168)
4™™™™™™™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™™™™™™ 3 2,268⫾79x 2,165⫾63x 1,997⫾57y 2.40⫾0.14x,y 2.59⫾0.11x 2.35⫾0.09y x,y x 1.94⫾0.11 2.01⫾0.08 1.76⫾0.07y 2.12⫾0.11x,y 2.17⫾0.08x 1.92⫾0.07y x x 6.4⫾0.2 6.0⫾0.2 5.5⫾0.2y 3.3⫾0.2x 3.0⫾0.2x,y 2.7⫾0.1y x x,y 30.6⫾0.4 30.0⫾0.3 29.4⫾0.3y 11.0⫾0.2x 10.7⫾0.1x 10.3⫾0.1y x,y x 14.4⫾0.2 14.7⫾0.2 14.0⫾0.1y 56.4⫾0.4x 56.8⫾0.3x 57.8⫾0.2y x x 1,163⫾44 1,132⫾34 974⫾30y 16.2⫾0.6x 15.8⫾0.4x 13.8⫾0.4y 9,526⫾500x,y 10,012⫾390x 8,660⫾346y 163.2⫾7.1x,y 166.2⫾5.1x 152.2⫾4.3y x x 1.88⫾0.06 1.84⫾0.05 1.62⫾0.04y 337.1⫾12.9x 334.7⫾9.4x 292.2⫾8.0y 12.4⫾0.4x 12.2⫾0.3x 10.6⫾0.2y
P valuea ⬍.001 .011 ⬍.001 .002 ⬍.001 .004 .006 ⬍.001 ⬍.001 .002 ⬍.001 ⬍.001 ⬍.001 .017 ⬍.001 ⬍.001 ⬍.001
a
P values are for transformed variables (except for macronutrients as % of energy and calcium intake). SE⫽standard error. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b
takes than boys not using weight-control behaviors or using only healthful behaviors (Table 3). Fruit intake was highest among boys reporting unhealthful weight-control behaviors and lowest among boys reporting no weightcontrol behaviors, being intermediate in boys using only healthful weight-control behaviors. Total fat and saturated fat intakes, as percentages of energy, were lower among boys using healthful and unhealthful weight-control behaviors than among boys not trying to control their weight. There were no differences in intakes of vegetables, protein, or carbohydrate (as percentages of energy intake), or any of the micronutrients examined (calcium; iron; vitamins A, C, and B-6; folate; and zinc) across weight-control behaviors. Similar analyses on boys were also run, adjusting for ethnicity/race, SES, and school level, and similar patterns were found (data not shown). Energy-adjusted analyses in boys further suggest that boys engaging in weight-control behaviors, particularly healthful behaviors, may have a higher quality of dietary intake than boys not engaging in weight-control behaviors (data not shown). Fruit and folate intake were higher in boys using healthful and unhealthful weight-control behaviors, and snack intake was lower, than in boys not using weight-control behaviors. Iron and vitamin B-6 intakes were higher in boys using healthful weight-control behaviors than in both of the other groups. Finally, zinc intake was higher in boys using healthful weight-control behaviors than in those using unhealthful weight-control behaviors. All other findings from energy-adjusted analyses were similar to those found in unadjusted analyses shown in Table 3.
916
June 2004 Volume 104 Number 6
Percentages of Youth Eating According to Dietary Guidelines by Weight-Control Behaviors Girls using unhealthful weight-control behaviors were less likely to meet dietary recommendations for most of the micronutrients examined (calcium, iron, vitamins C and B-6, folate, and zinc) than girls using no weightcontrol behaviors or only healthful behaviors. In addition, girls using unhealthful weight-control behaviors were less likely to consume the recommended amounts of fruits, vegetables, and grains than girls using healthful weight-control behaviors (Table 4). Only one fourth (25.7%) of girls using unhealthful weight-control behaviors consumed 1,300 mg/day or more of calcium, and only one third (33.4%) met dietary recommendations for iron intake. Analyses adjusting for race/ethnicity, SES, and school level revealed similar patterns (data not shown). Compared with girls, among boys there were fewer differences across weight-control behaviors, and differences were not consistent in their direction (Table 5). For example, boys using unhealthful weight-control behaviors were less likely to meet dietary recommendations for iron but were more likely to meet recommendations for fruit, vegetable, and saturated fat intakes than boys not engaging in weight-control behaviors. Among the boys, in analyses adjusting for race/ethnicity, SES, and school level, patterns were similar to those shown in Table 5 except for fruit, iron, and total fat intakes. For fruit and iron intake, differences across weight-control behaviors were no longer statistically significant. But, significantly more boys using unhealthful
Table 3. Dietary intake (mean/day) by types of weight-control behaviors used in the past year in adolescent boys Weight-Control Behaviors None (Nⴝ610) Energy (kcal) Fruits (servings) Vegetables (servings not including french fries) Vegetables (servings including french fries) Grains (servings) Sweet/salty snack foods (servings) Total fat (% of energy) Saturated fat (% of energy) Protein (% of energy) Carbohydrate (% of energy) Calcium (mg) Iron (mg) Vitamin A (IU) Vitamin C (mg) Vitamin B-6 (mg) Folate (g) Zinc (mg)
Healthful only (Nⴝ798)
Unhealthful (Nⴝ650)
4™™™™™™™™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™™™3 2,301⫾72 2,238⫾68 2,289⫾69 2.12⫾0.09x 2.32⫾0.08x,y 2.55⫾0.08y 1.61⫾0.08 1.64⫾0.07 1.76⫾0.08 1.80⫾0.09 1.80⫾0.08 1.95⫾0.08 6.3⫾0.2 6.3⫾0.2 6.3⫾0.2 3.3⫾0.2x 2.9⫾0.2y 3.1⫾0.2x,y 31.0⫾0.2x 30.3⫾0.2y 30.0⫾0.2y 11.3⫾0.1x 10.7⫾0.1y 10.6⫾0.1y 14.6⫾0.1 14.9⫾0.1 14.6⫾0.1 55.4⫾0.3 55.9⫾0.2 56.3⫾0.3 1,222⫾41 1,207⫾38 1,148⫾39 15.8⫾0.6 16.0⫾0.5 15.6⫾0.6 8,659⫾334 9,041⫾303 8,871⫾318 152.7⫾5.1 155.3⫾4.6 162.9⫾4.9 1.85⫾0.05 1.87⫾0.05 1.81⫾0.05 332.8⫾10.9 338.2⫾10.0 325.0⫾10.4 12.7⫾0.4 12.8⫾0.4 12.3⫾0.4
P valuea .566 .002 .401 .114 .725 .006 .003 ⬍.001 .059 .054 .107 .253 .361 .177 .349 .230 .392
a
P values are for transformed variables (except for macronutrients as % of energy and calcium intake). SE⫽standard error. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b
weight-control behaviors met dietary recommendations for total fat intake than boys reporting no weight-control behaviors (P⫽.027).
Interactions with Ethnicity/Race and Weight Status In general, ethnicity/race did not confound the relationship between dietary intake and weight-control behav-
Table 4. Percentage of adolescent girls meeting dietary recommendationsa by types of weight-control behaviors Weight-Control Behaviors
Fruits Vegetables (not including french fries) Vegetables (including french fries) Grains Total fat Saturated fat Calcium Iron Vitamin A Vitamin C Vitamin B-6 Folate Zinc
Dietary recommendation
None (Nⴝ253)
Healthful only (Nⴝ622)
Unhealthful (Nⴝ1,168)
ⱖ2 servings/day ⱖ3 servings/dayc ⱖ3 servings/day ⱖ6 servings/day ⱕ30% of energy intake ⱕ10% of energy intake ⱖ1,300 mg/day ⱖ15 mg/day ⱖ2,330 IU/day ⱖ65 mg/day ⱖ1.2 mg/day ⱖ400 g/day ⱖ9 mg/day
4™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™™3 46.8⫾3.5x,y 50.5⫾2.5x 41.2⫾2.1y x,y x 16.2⫾2.5 20.1⫾1.8 14.0⫾1.5y 17.2⫾2.6x,y 22.8⫾1.9y 16.2⫾1.6x x x 45.4⫾3.3 40.2⫾2.4 34.3⫾2.0y 47.2⫾3.6 48.9⫾2.6 51.5⫾2.2 31.6⫾3.7x 40.4⫾2.8y 45.0⫾2.5y 39.6⫾3.2x 33.6⫾2.4x 25.7⫾2.1y x x 45.7⫾3.3 42.5⫾2.3 33.4⫾1.9y 92.6⫾1.9 93.1⫾1.3 89.9⫾1.0 83.9⫾2.6x 83.1⫾1.7x,y 77.9⫾1.3y 76.4⫾3.2x 72.7⫾2.2x 63.0⫾1.8y x x 29.8⫾3.0 30.1⫾2.1 22.6⫾1.8y 66.9⫾3.3x 63.7⫾2.3x 52.4⫾1.8y
P value ⬍.001 .003 .001 .001 .356 ⬍.001 ⬍.001 .001 .049 .012 ⬍.001 ⬍.001 ⬍.001
a
Dietary recommendations are Healthy People 2010 Nutrition Objectives for fruits and vegetables and Dietary Reference Intakes for nutrients. SE⫽standard error. The recommended amount does not exclude french fries; therefore, comparisons should be made cautiously. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b c
Journal of THE AMERICAN DIETETIC ASSOCIATION
917
Table 5. Percentage of adolescent boys meeting dietary recommendationsa by types of weight-control behaviors
Fruits Vegetables (not including french fries) Vegetables (including french fries) Grains Total fat Saturated fat Calcium Iron Vitamin A Vitamin C Vitamin B-6 Folate Zinc
Weight-Control Behaviors
Dietary recommendation
None
ⱖ2 servings/day ⱖ3 servings/dayc ⱖ3 servings/day ⱖ6 servings/day ⱕ30% of energy intake ⱕ10% of energy intake ⱖ1,300 mg/day ⱖ11 mg/day ⱖ3,000 IU/day ⱖ75 mg/day ⱖ1.3 mg/day ⱖ400 g/day ⱖ11 mg/day
4™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™3 41.8⫾2.6x 44.6⫾2.3x,y 48.9⫾2.5y x x,y 11.4⫾1.8 12.9⫾1.6 16.6⫾1.7y 13.9⫾1.9x 15.1⫾1.7x,y 18.9⫾1.8y 45.6⫾2.7 42.8⫾2.4 40.0⫾2.5 41.2⫾2.0 45.2⫾1.8 47.8⫾2.0 28.0⫾2.3x 37.2⫾2.0y 40.7⫾2.2y 42.0⫾2.7 40.9⫾2.5 38.8⫾2.6 67.1⫾2.8x 67.1⫾2.6x 59.9⫾2.7y 87.8⫾1.7 88.8⫾1.5 85.8⫾1.6 77.2⫾1.8 78.4⫾1.6 74.9⫾1.7 68.3⫾2.3 69.7⫾2.1 64.4⫾2.2 29.9⫾2.3 32.8⫾0.2 30.1⫾2.2 54.0⫾2.9 52.8⫾2.7 47.5⫾2.8
Healthful only
Unhealthful
P value .045 .020 .035 .151 .072 ⬍.001 .525 .007 .229 .312 .101 .419 .053
a
Dietary recommendations are Healthy People 2010 Nutrition Objectives for fruits and vegetables and Dietary Reference Intakes for nutrients. SE⫽standard error. c The recommended amount does not exclude french fries; therefore, comparisons should be made cautiously. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b
iors. The only exception was that African-American girls using healthful weight-control behaviors were more likely to meet dietary recommendations for vegetables (30.8%) than white girls using healthful weight-control behaviors (14.2%) (overall P⫽.007 for 8 df test, post hoc P⫽.009 for 2 df test). In contrast, vegetable intake did not differ between other African-American and white girls. Although the P values are small, inferences must be made cautiously because many outcomes were tested for interactions with ethnicity/race, and the post hoc tests were not preplanned, but rather were data driven. Overweight status did not confound any of the relationships between dietary intake and weight-control behavior in either girls or boys. DISCUSSION The current study explored associations between different types of weight-control behaviors and dietary intake among adolescent girls and boys. Among adolescent girls, the patterns of association between unhealthful weight behaviors and dietary intake were of greater concern than among the boys. Girls engaging in unhealthful weight-control behaviors had lower intakes of fruits, vegetables, grains, and all of the micronutrients assessed in this study than girls reporting only healthful weightcontrol behaviors. As compared with girls not reporting any weight-control behaviors, girls engaging in unhealthful weight-control behaviors had lower intakes of most of the micronutrients that were examined. Of particular concern were the low intakes of calcium and iron in girls using unhealthful weight-control behaviors, because intake of these nutrients tends to be less than dietary recommendations for all adolescent girls, regardless of their dieting status (33). In contrast to the girls, boys reporting weight-control
918
June 2004 Volume 104 Number 6
behaviors (healthful and unhealthful) did not differ greatly from boys not reporting any weight-control behaviors. Furthermore, the few differences that were found in the boys suggested an improved dietary intake among boys trying to control their weight. Sex differences in patterns of association between unhealthful weight-control behaviors and dietary intake found in the current study suggest that the use of unhealthful weight-control behaviors represents a different set of behaviors (and possibly associated attitudes) for boys than for girls. Further research is needed to better understand motivations and behaviors of adolescent boys reporting the use of healthful and unhealthful weight-control behaviors. Findings from the current study confirm previous findings regarding sex differences and the importance of distinguishing between healthful and unhealthful weightcontrol behaviors. In a study using data from the Youth Risk Behavior Survey, extremely unhealthful weightcontrol behaviors (defined as vomiting or diet pill use) were associated with decreased fruit and vegetable intake in girls. In contrast, moderate weight-control behaviors (defined as all other methods) were associated with increased fruit and vegetable intake and decreased highfat food intake in both girls and boys (15). The high prevalence of girls using unhealthful weightcontrol behaviors (64% of the girls) and the strong associations between these behaviors and inadequate dietary intake point to a need for education about healthful strategies for adolescent girls who are interested in weight management. It may be helpful to deter girls from “dieting,” which tends to be short-term, and encourage adoption of more healthful behaviors that can be integrated into one’s lifestyle, such as substituting fruits and vegetables for higher-fat snack foods. Findings from ethnographic research done by Nichter and colleagues (34) on
adolescent girls suggests differences between “dieting” and “watching what I eat.” “Dieting” tended to be associated with poorer dietary intake, whereas “watching” was actually associated with higher calcium intake. It may be important to teach adolescents how to “watch” what they eat, to ensure that if energy intake is reduced, nutrient intake remains stable and adequate. That said, it is important to note that even the behaviors classified in this paper as “healthful” may turn into unhealthful behaviors if done excessively. Thus, any messages promoting the use of healthful weight-control behaviors need to be given with caution. Strengths of the present study include the comprehensive assessments of weight-control behaviors and dietary intake. Most large population-based studies of youth address a broader range of adolescent health concerns and thus have only included a few questions about weightcontrol behaviors and dietary intake (15,16,21). In addition, although the study population was drawn from only one Midwestern state, its large and diverse nature in terms of ethnicity and socioeconomic backgrounds provides some confidence in making extrapolations to a broader sector of the adolescent population. However, study limitations also need to be considered. Although the time periods referred to in the questions on weightcontrol behaviors and dietary intake were similar (previous year), we cannot be sure that the behaviors were occurring simultaneously and that dietary intake was poorer during the exact time period that girls were engaging in unhealthful weight-control behaviors. Furthermore, we cannot assert that the relationship was causal and that the use of unhealthful weight-control behaviors was leading to poorer dietary intake. CONCLUSIONS Findings from the present study provide justification for interventions aimed at preventing unhealthful weightcontrol behaviors, particularly among adolescent girls. Interventions aimed at preventing unhealthful weightcontrol behaviors should include both educational and environmental components. Educational messages for both sexes can address the reasons why adolescents engage in unhealthful weight-control behaviors, the potential dangers and futility of unhealthful dieting, knowledge and skills for the use of healthful weightcontrol behaviors (when appropriate), and skills for resisting potentially harmful social influences to engage in unhealthful weight-control behaviors (35-37). Interventions should also directly address the social environments of the adolescent, including proximal factors such as weight-related norms and behaviors of family members and peers, and more distal factors, such as messages in teen magazines and other media regarding weight-control strategies, because all of these social sources influence whether and how an adolescent will try to control weight (38,39). Dietitians can play a major role in the prevention of unhealthful weight-control behaviors through their clinical work with adolescents and their families, outreach to schools and community centers, and advocacy work such as writing teen-friendly magazine articles on nutrition and weight control. Clearly the messages should focus on healthful methods to control weight and yet maintain
adequate nutrient intake. Dietitians are in a unique position in that they also have the training and skills to help adolescents at risk for obesity incorporate healthful weight-control behaviors into their lifestyle, whether for weight maintenance or gradual weight loss. References 1. Story M, French S, Resnick M, Blum R. Ethnic/racial and socioeconomic differences in dieting behaviors and body image perceptions in adolescents. Int J Eat Disord. 1995;18:173-179. 2. Serdula MK, Collins E, Williamson DF, Anda RF, Pamuk E, Byers TE. Weight control practices of U.S. adolescents and adults. Ann Intern Med. 1993;119: 667-671. 3. Neumark-Sztainer D, Hannan PJ. Weight-related behaviors among adolescent girls and boys: Results from a national survey. Arch Pediatr Adolesc Med. 2000;154:569-577. 4. Neumark-Sztainer D, Story M, Hannan PJ, Perry CL, Irving LM. Weight-related concerns and behaviors among overweight and non-overweight adolescents: Implications for preventing weight-related disorders. Arch Pediatr Adolesc Med. 2002;156:171-178. 5. Neumark-Sztainer D, Croll J, Story M, Hannan PJ, French S, Perry C. Ethnic/racial differences in weight-related concerns and behaviors among adolescent girls and boys: Findings from Project EAT. J Psychosom Res. 2002;53:963-974. 6. Spear BA. Adolescent nutrition: General, In: Worthington-Roberts BS, Williams SR, eds. Nutrition Throughout the Life Cycle. 4th ed. Boston: McGrawHill Higher Education; 2000:262-287. 7. Gong EJ, Spear BA. Adolescent growth and development: Implications for nutritional needs. J Nutr Educ. 1988;20:273-279. 8. Lytle LA. Nutritional issues for adolescents. J Am Diet Assoc. 2002;102(3 Suppl):S8-S12. 9. Spear BA. Adolescent growth and development. J Am Diet Assoc. 2002;102(3 Suppl):S23-S29. 10. Optimal calcium intake. NIH Consensus Statement Online 1994. NIH Consensus Development Conference, June 6-8, 1994; 12(4):1-31. Available at: http:// odp.od.nih.gov/Consensus/cons/097/097_statement. htm. Accessed September 25, 2003. 11. Centers for Disease Control and Prevention. Recommendations to prevent and control iron deficiency in the United States. Mor Mortal Wkly Rep CDC Surveill Summ. 1998;Apr 3:1-29. 12. Macdonald LA, Wearring GA, Moase O. Factors affecting the dietary quality of adolescent girls. J Am Diet Assoc. 1983;82:260-263. 13. Gibbons KL, Wertheim EH, Paxton SJ, Petrovich J, Szmukler GI. Nutrient intake of adolescents and its relationship to desire for thinness, weight loss behaviours, and bulimic tendencies. Aust J Nutr Diet. 1995; 52:69-74. 14. Barber SA, Bull NL. Food and nutrient intakes by British women aged 15-25 years, with particular reference to dieting habits and iron intakes. Ecol Food Nutr. 1985;16:161-169. 15. Story M, Neumark-Sztainer D, Sherwood N, Stang J, Murray D. Dieting status and its relationship to eat-
Journal of THE AMERICAN DIETETIC ASSOCIATION
919
16.
17. 18. 19.
20.
21.
22.
23.
24.
25.
26.
27.
920
ing and physical activity behaviors in a representative sample of U.S. adolescents. J Am Diet Assoc. 1998;98:1127-1135, 1255. Neumark-Sztainer D, Story M, Dixon LB, Resnick M, Blum R. Correlates of inadequate consumption of dairy products among adolescents. J Nutr Educ. 1997;29:12-20. Crawley H, Shergill-Bonner R. The nutrient and food intakes of 16-17 year old female dieters in the UK. J Hum Nutr. 1995;8:25-34. Barr SI. Dieting attitudes and behavior in urban high school students: Implications for calcium intake. J Adolesc Health. 1995;16:458-464. Webster BL, Barr SI. Calcium intakes of adolescent female gymnasts and speed skaters: Lack of association with dieting behavior. Int J Sport Nutr. 1995;5: 2-12. Mulvihill CB, Davies GJ, Rogers PJ. Dietary restraint in relation to nutrient intake, physical activity and iron status in adolescent females. J Hum Nutr Diet. 2002;15:19-31. Middleman AB, Vazquez I, Durant RH. Eating patterns, physical activity, and attempts to change weight among adolescents. J Adolesc Health. 1998; 22:37-42. Brugman E, Meulmeester JF, Spee-van der Wekke A, Beuker RJ, Zaadstra BM, Radder JJ, Verloove-Vanhorick PS. Dieting, weight and health in adolescents in The Netherlands. Int J Obesity. 1997;21:54-60. Neumark-Sztainer D, Story M, Resnick MD, Blum RW. Correlates of inadequate fruit and vegetable consumption among adolescents. Prev Med. 1996;25: 497-505. Neumark-Sztainer D, Story M, Hannan P, Moe J. Overweight status and eating patterns among adolescents: Where do youth stand in comparison to the Healthy People 2010 Objectives? Am J Public Health. 2002;92:844-851. Neumark-Sztainer D, Wall M, Story M, Perry CL. Correlates of unhealthful weight-control behaviors among adolescent girls and boys: Implications for the primary prevention of disordered eating. Health Psychol. 2003;22:88-98. Neumark-Sztainer D, Hannan PJ, Story M, Croll J, Perry C. Family meal patterns: Associations with sociodemographic characteristics and improved dietary intake among adolescents. J Am Diet Assoc. 2003;103:317-322. Rockett HRH, Breitenbach MA, Frazier AL, Witschi J, Wolf AM, Field AE, Colditz GA. Validation of a youth/adolescent food frequency questionnaire. Prev Med. 1997;26:808-816.
June 2004 Volume 104 Number 6
28. Rockett HR, Wolf AM, Colditz GA. Development and reproducibility of a food frequency questionnaire to assess diets of older children and adolescents. J Am Diet Assoc. 1995;95:336-340. 29. Healthy People 2010. Washington, DC: US Dept of Health and Human Services; 2000. 30. Yates AA, Schlicker SA, Suitor CW. Dietary Reference Intakes: The new basis for recommendations for calcium and related nutrients, B vitamins, and choline. J Am Diet Assoc. 1998;98:699-706. 31. Food and Nutrition Board, Institute of Medicine, National Academy of Sciences. New dietary reference intakes: Recommended levels for individual intakes 1997 and 1998, recommended daily allowances, revised 1989. Nutr Today. 1998;33:257-260. 32. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board, Institute of Medicine. New dietary reference intakes for calcium and related nutrients. Nutrition Research Newsletter. 1997;16:85-87. 33. Alaimo K, McDowell MA, Briefel RR, Bischof AM, Caughman CR, Loria CM, Johnson CL. Dietary intake of vitamins, minerals, and fiber of persons ages 2 months and over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988-91. Washington, DC: National Center for Health Statistics; 1994. Advance Data from Vital and Health Statistics. 258:1-28. 34. Nichter M, Ritenbaugh C, Nichter M, Vuckovic N, Aickin M. Dieting and “watching” behaviors among adolescent females: Report of a multimethod study. J Adolesc Health. 1995;17:153-162. 35. Neumark-Sztainer D, Story M, Hannan PJ, Rex J. New Moves: A school-based obesity prevention program for adolescent girls. Prev Med. 2003;37:41-51. 36. Littleton HL, Ollendick T. Negative body image and disordered eating behavior in children and adolescents: What places youth at risk and how can these problems be prevented? Clin Child Fam Psychol Rev. 2003;6:51-66. 37. Paxton SJ, Wertheim EH, Pilawski A, Durkin S, Holt T. Evaluations of dieting prevention messages by adolescent girls. Prev Med. 2002;35:474-491. 38. Martinez-Gonzalez MA, Gual P, Lahortiga F, Alonso Y, de Irala-Estevez J, Cervera S. Parental factors, mass media influences, and the onset of eating disorders in a prospective population-based cohort. Pediatrics. 2003;111:315-320. 39. Utter J, Neumark-Sztainer D, Wall M, Story M. Reading magazine articles about dieting and associated weight-control behaviors among adolescents. J Adolesc Health. 2003;32:78-82.
Weight-Control Behaviors among Adolescent Girls and Boys: Implications for Dietary Intake DIANNE NEUMARK-SZTAINER, PhD, MPH, RD; PETER J. HANNAN, MStat; MARY STORY, PhD, RD; CHERYL L. PERRY, PhD
ABSTRACT Objective To examine associations between healthful and unhealthful weight-control behaviors and dietary intake among adolescents. Design Cross-sectional survey (Project EAT [Eating Among Teens]). Subjects/Setting The study population included 4,144 middle and high school students from Minneapolis/St. Paul public schools from diverse racial and socioeconomic backgrounds. Statistical Analyses Dietary intake patterns were compared across adolescent girls and boys reporting unhealthful, only healthful, or no weight-control behaviors in unadjusted analyses and analyses adjusted for sociodemographic factors and energy intake. Results Among girls, mean intakes differed across weightcontrol behaviors for all foods and nutrients examined. P values ranged from P⫽.006 to P⬍.001. Girls using unhealthful weight-control behaviors had significantly lower intakes of fruit; vegetables; grains; calcium; iron; vitamins A, C, and B-6; folate; and zinc than girls using only healthful weight-control behaviors. Compared with girls reporting no weight-control behaviors, girls using unhealthful weight-control behaviors had lower intakes of grains, calcium, iron, vitamin B-6, folate, and zinc. In contrast to the girls, boys reporting unhealthful weightcontrol behaviors did not have poorer dietary intakes than boys not using weight-control behaviors or using only healthful behaviors. Among boys, there were no significant differences in mean intakes of vegetables; grains; calcium; iron; vitamins A, C, and B-6; folate; and zinc. Furthermore, boys using unhealthful weight-control behaviors had higher fruit intakes (P⫽.002) than boys reporting no weight-control behaviors. Conclusions Adolescent girls who engage in unhealthful weight-control behaviors are at increased risk for dietary inadequacy. The findings demonstrate a need for interventions to prevent unhealthful weight-control behaviors D. Neumark-Sztainer, P. J. Hannan, M. Story, and C. L. Perry are with the Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis. Address correspondence to: Dianne Neumark-Sztainer, PhD, MPH, RD, Division of Epidemiology, School of Public Health, University of Minnesota, 1300 S Second St, Suite 300, Minneapolis, MN 55454. E-mail: Neumark@ epi.umn.edu Copyright © 2004 by the American Dietetic Association. 0002-8223/04/10406-0006$30.00/0 doi: 10.1016/j.jada.2004.03.021
© 2004 by the American Dietetic Association
in adolescent girls, and to promote healthful weight-control behaviors when indicated. J Am Diet Assoc. 2004;104:913-920.
D
ata from numerous studies have demonstrated a high prevalence of weight-control behaviors among adolescents, particularly adolescent girls (1-3). In previously reported findings from the Project EAT (Eating Among Teens) study, 45% of adolescent girls and 21% of adolescent boys reported that they were currently trying to lose weight, and an additional 26% of girls and 23% of boys were trying to maintain their weight (4,5). Of particular concern were the high percentages of youth who reported using unhealthful weight-control behaviors. Unhealthful weight-control behaviors such as skipping meals, fasting, using food substitutes, or smoking more cigarettes were reported by more than half (57%) of the girls and by one third (33%) of the boys. An additional 12% of girls and 5% of boys reported extremely unhealthful behaviors such as use of diet pills, laxatives, diuretics, or vomiting. Some weight-control behaviors (eg, increasing fruits and vegetables and decreasing foods high in fats and sugars) are appropriate, even desirable, for most adolescents, when done in moderation. These behaviors have the potential to help adolescents achieve a healthful body weight through the prevention of excess weight gain or, when appropriate, through gradual weight loss. However, because of the importance of adequate nutrition during adolescence (6-9), it is essential to determine the potential implications of different weight-control behaviors on dietary intake in youth. On one hand, there may be positive implications if adolescents become more attentive eaters, consume more fruits, vegetables, and whole grains, and avoid excessive fat intake. On the other hand, there may be negative implications if energy restrictions or selective food choices lead to decreased intake of nutrients such as iron and calcium, which are important during this period of growth and development (10,11). To guide the development of appropriate messages and dietary intervention programs for youth, it is important to determine whether weight-control behaviors are associated with dietary intake and whether associations vary for different types of weight-control behaviors. In general, studies have suggested that youth engaging in weight-control behaviors (unhealthful weight-control behaviors in particular), are at risk for inadequate dietary intakes (12-17), although in some studies associations were not found (18-20), and in other studies weightcontrol behaviors were associated with more positive eating patterns such as decreased consumption of high-
Journal of THE AMERICAN DIETETIC ASSOCIATION
913
fat baked goods (ie, cookies, doughnuts, pie, or cake) (21), sweets, salty snacks, and soft drinks (22). Inconsistencies may be due to differences in measures used to assess weight-control behaviors, dietary assessment tools, dietary components being examined, and sample characteristics and sizes. Many of the studies that have examined associations between weight-control behaviors and dietary intake did not distinguish between different types of weight-control behaviors (16,18,21,23), used inadequate measures of dietary intake (15,16,21-23), had small samples (or small numbers of dieters within samples) (12-14,19), or included only girls in their study populations (12,17,20). The current study explores associations between weight-control behaviors and dietary intake. It expands on existing studies exploring these associations in that data are drawn from a large, ethnically diverse population of adolescents. In addition, healthful weight-control behaviors, unhealthful weight-control behaviors, and a broad array of dietary outcomes are examined. Specifically, the current study compares intake patterns of key foods and nutrients among adolescent girls and boys reporting no weight-control behaviors in the past year, only healthful weight-related behaviors, or unhealthful weight-control behaviors. Mean dietary intakes and percentages of adolescent girls and boys meeting dietary recommendations are examined across these three groups. The current study further explores whether associations between weight-control behaviors and dietary intake differ across race/ethnicity and weight status of adolescent girls and boys. METHODS Study Population and Study Design The overall study population included 4,746 adolescents from 31 public middle schools and high schools from urban and suburban school districts in the St Paul/Minneapolis area of Minnesota who participated in Project EAT. Trained research staff administered surveys within school classes and assessed height and weight within a private area. Study procedures were approved by the University of Minnesota Human Subjects Committee and by the research boards of the participating school districts. Consent procedures were done in accordance with the requests of the participating school districts; in some schools passive consent procedures were utilized, whereas active consent procedures were required in others. The response rate for student participation was 81.5%. Further details on the study sample and design have been previously published (24-26). Participants who did not complete the food frequency questionnaire (n⫽344) were not included in the current analysis. In addition, participants who had energy intakes less than 400 kcal/day or more than 7,000 kcal/day (n⫽258) were excluded from analyses examining associations with dietary intake because these values are considered biologically implausible for habitual intake. Thus, the final study population in the current analysis included 4,144 participants. The adolescents included in the current analysis were compared with those who were excluded due to missing or unusable dietary intake data and found to be similar in
914
June 2004 Volume 104 Number 6
terms of sex and body mass index (BMI), but more likely to be white, of higher socioeconomic status, and in middle school (rather than high school). These differences were not expected to strongly influence associations between weight control practices and dietary intake that are being explored in the analyses. Measures Weight-control behaviors over the past year were assessed with the question: “Have you done any of the following things in order to lose weight or keep from gaining weight during the past year?” (yes/no for each method). Responses classified as healthful weight-control behaviors included lifestyle behaviors recommended for all adolescents for overall health promotion and weight management: (a) exercise, (b) ate more fruits and vegetables, (c) ate less high-fat foods, and (d) ate less sweets. Responses classified as unhealthful weight-control behaviors included: (a) fasted, (b) ate very little food, (c) used food substitute (powder/special drink), (d) skipped meals, (e) smoked more cigarettes, (f) took diet pills, (g) made myself vomit, (h) used laxatives, and (i) used diuretics. Respondents were classified as using healthful weight-control behaviors if they reported any of the healthful behaviors and no unhealthful behaviors. They were categorized as using unhealthful weight-control behaviors if they reported any of the unhealthful behaviors. Most of the youth using unhealthful weight-control behaviors also reported the use of healthful weight-control behaviors; therefore, mutually exclusive groups were not possible. For the majority of specific weight-control behaviors used in the past year, test-retest values over a 2-week interval ranged from 0.50 to 0.68, but lower values were found for laxatives (0.29) and food substitutes (0.44). Dietary intake was assessed with the Youth and Adolescent Food Frequency Questionnaire (27,28). Servings of fruits, vegetables (including and not including french fries), grains (eg, cereal, bread, pasta, rice), and sweet and salty snack foods (eg, potato chips, candy, cake, and cookies) were examined in addition to the following nutrients: energy (kcal), total fat (% of total energy), saturated fat (% of total energy), protein (% of total energy), carbohydrate (% of total energy), calcium (mg), iron (mg), vitamin A (IU), vitamin C (mg), vitamin B-6 (mg), folate (g), and zinc (mg). In addition to examining mean intakes, we also examined the percentages of youth meeting the Healthy People 2010 objectives for fruits, vegetables, and grains (29), and the Dietary Reference Intakes (DRI) for macronutrients and micronutrients (30-32). Sex, school level, ethnicity/race, and socioeconomic status (SES) were based on self-report (24). BMI was based on height and weight measurements taken by research staff using standardized equipment and procedures. Data Analysis Girls and boys were analyzed separately due to sex differences in weight-control behaviors. Analyses of variance generated mean outcomes for adolescents reporting no weight-control behaviors, only healthful weight-control behaviors, or unhealthful weight-control behaviors.
To allow for the clustered design, the school was included in all analyses as a random effect. For skewed outcomes such as servings of fruits and vegetables, the square root of the outcome was analyzed; however, means are reported on the natural scale for all outcomes. The percent of students meeting the various Year 2010 objectives were treated as continuous measures. Analysis was by PROC MIXED (SAS, Release 8.2). Covariates used in adjusted analyses were grade (middle vs high school), ethnicity/race (white, African American, Hispanic, Asian American, other), and SES (five levels). Analyses examining mean intakes of foods and nutrients were further adjusted for energy intake, but analyses examining the percentages of adolescents eating according to dietary recommendations were not energyadjusted because most dietary recommendations are consistent across energy intake. Further energy adjustments for macronutrients were not done because they are reported as percentages of total energy intake in all analyses. Results from unadjusted analyses are presented in the tables; findings from adjusted analyses are discussed in the text. The possibility of confounding of the relationship between dietary intake and weight control methods by race/ ethnicity was examined by including the interaction of race/ethnicity (five categories) with weight-control behavior (three categories) in the adjusted analyses. A significance level of less than .01 was required for the global F-test [8 degrees of freedom (df)]. Where the global test was significant, more specific post hoc hypotheses were examined comparing each of the four minority racial/ ethnic groups against the majority white group (2 df). Adjusted means were generated to describe the pattern of interaction. For weight status as a possible confounder, overweight was defined as BMI at or more than the 85th percentile, and the main effect for overweight and interaction of overweight with weight-control methods were added to the adjusted model. Again, a significance level less than .01 was required for the global test of interaction (2 df test). RESULTS Description of Study Population Table 1 provides a description of the study population in terms of age, BMI, ethnicity/race, SES, and weight-control behaviors. More than half of the girls (57.2%) and nearly one third of the boys (31.6%) reported the use of at least one unhealthful weight-control behavior over the past year. Mean Dietary Intake by Weight-Control Behaviors In general, girls using unhealthful weight-control behaviors had poorer dietary intakes than girls reporting no weight-control behaviors or only healthful behaviors (Table 2). Girls using unhealthful weight-control behaviors had significantly lower intakes of fruit, vegetables, grains, and all micronutrients examined (calcium; iron; vitamins A, C, and B-6; folate; and zinc) than girls using only healthful weight-control behaviors. Compared with girls reporting no weight-control behaviors, girls using unhealthful weight-control behaviors had lower intakes
Table 1. Description of study population: Body mass index, sociodemographic characteristics, and weight-control behaviors Adolescent girls (Nⴝ2,064a) Age (y) BMIc Race/ethnicity White African American Hispanic Asian American Native American Mixed/other SESd High High-middle Middle Low-middle Low Grade level Middle school High school Weight-control behaviors None Healthful only Unhealthful
Adolescent boys (Nⴝ2,080a)
Mean 14.8 23.2
SDb 1.67 4.94
N
%
N
%
997 340 100 426 85 86
49.1 16.7 4.9 20.9 4.2 4.2
1,120 316 123 377 58 61
54.4 15.4 6.0 18.4 2.8 3.0
274 448 519 382 386
13.7 22.3 25.8 19.0 19.2
297 528 552 372 278
14.7 26.0 27.2 18.4 13.7
680 1,358
33.4 66.6
675 1,385
32.8 67.2
253 622 1,168
12.4 30.5 57.1
610 798 650
29.6 38.8 31.6
Mean 14.9 23.0
SD 1.65 4.81
a Due to missing values for certain variables, numbers do not always add up to this maximum N. b SD⫽standard deviation. c BMI⫽body mass index. d SES⫽socioeconomic status.
of grains and certain micronutrients (calcium, iron, vitamin B-6, folate, and zinc). Analyses adjusting for ethnicity/race, SES, and school level found similar patterns (data not shown in tables). In energy-adjusted analyses, differences between girls using unhealthful and healthful weight-control behaviors remained statistically significant for calcium, iron, vitamins A and B-6, folate, and zinc, with lower intakes among girls using unhealthful behaviors, but differences between girls using unhealthful behaviors and no weightcontrol behaviors were no longer statistically significant for these nutrients (data not shown). Girls using only healthful weight-control behaviors had higher fruit intake than girls not using weight-control behaviors and higher vitamin A intake than both of the other groups. Associations between weight-control behaviors and servings of vegetables, grains, snack foods, and vitamin C were no longer statistically significant. Other findings from energy-adjusted analyses were similar to those shown in Table 2 for unadjusted analyses. In contrast to the girls, boys reporting unhealthful weight-control behaviors did not have poorer dietary in-
Journal of THE AMERICAN DIETETIC ASSOCIATION
915
Table 2. Dietary intake (mean/day) by types of weight-control behaviors used in the past year in adolescent girls Weight-Control Behaviors None (Nⴝ253) Energy (kcal) Fruits (servings) Vegetables (servings not including french fries) Vegetables (servings including french fries) Grains (servings) Sweet/salty snack foods (servings) Total fat (% of energy) Saturated fat (% of energy) Protein (% of energy) Carbohydrate (% of energy) Calcium (mg) Iron (mg) Vitamin A (IU) Vitamin C (mg) Vitamin B-6 (mg) Folate (g) Zinc (mg)
Healthful only (Nⴝ622)
Unhealthful (Nⴝ1,168)
4™™™™™™™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™™™™™™ 3 2,268⫾79x 2,165⫾63x 1,997⫾57y 2.40⫾0.14x,y 2.59⫾0.11x 2.35⫾0.09y x,y x 1.94⫾0.11 2.01⫾0.08 1.76⫾0.07y 2.12⫾0.11x,y 2.17⫾0.08x 1.92⫾0.07y x x 6.4⫾0.2 6.0⫾0.2 5.5⫾0.2y 3.3⫾0.2x 3.0⫾0.2x,y 2.7⫾0.1y x x,y 30.6⫾0.4 30.0⫾0.3 29.4⫾0.3y 11.0⫾0.2x 10.7⫾0.1x 10.3⫾0.1y x,y x 14.4⫾0.2 14.7⫾0.2 14.0⫾0.1y 56.4⫾0.4x 56.8⫾0.3x 57.8⫾0.2y x x 1,163⫾44 1,132⫾34 974⫾30y 16.2⫾0.6x 15.8⫾0.4x 13.8⫾0.4y 9,526⫾500x,y 10,012⫾390x 8,660⫾346y 163.2⫾7.1x,y 166.2⫾5.1x 152.2⫾4.3y x x 1.88⫾0.06 1.84⫾0.05 1.62⫾0.04y 337.1⫾12.9x 334.7⫾9.4x 292.2⫾8.0y 12.4⫾0.4x 12.2⫾0.3x 10.6⫾0.2y
P valuea ⬍.001 .011 ⬍.001 .002 ⬍.001 .004 .006 ⬍.001 ⬍.001 .002 ⬍.001 ⬍.001 ⬍.001 .017 ⬍.001 ⬍.001 ⬍.001
a
P values are for transformed variables (except for macronutrients as % of energy and calcium intake). SE⫽standard error. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b
takes than boys not using weight-control behaviors or using only healthful behaviors (Table 3). Fruit intake was highest among boys reporting unhealthful weight-control behaviors and lowest among boys reporting no weightcontrol behaviors, being intermediate in boys using only healthful weight-control behaviors. Total fat and saturated fat intakes, as percentages of energy, were lower among boys using healthful and unhealthful weight-control behaviors than among boys not trying to control their weight. There were no differences in intakes of vegetables, protein, or carbohydrate (as percentages of energy intake), or any of the micronutrients examined (calcium; iron; vitamins A, C, and B-6; folate; and zinc) across weight-control behaviors. Similar analyses on boys were also run, adjusting for ethnicity/race, SES, and school level, and similar patterns were found (data not shown). Energy-adjusted analyses in boys further suggest that boys engaging in weight-control behaviors, particularly healthful behaviors, may have a higher quality of dietary intake than boys not engaging in weight-control behaviors (data not shown). Fruit and folate intake were higher in boys using healthful and unhealthful weight-control behaviors, and snack intake was lower, than in boys not using weight-control behaviors. Iron and vitamin B-6 intakes were higher in boys using healthful weight-control behaviors than in both of the other groups. Finally, zinc intake was higher in boys using healthful weight-control behaviors than in those using unhealthful weight-control behaviors. All other findings from energy-adjusted analyses were similar to those found in unadjusted analyses shown in Table 3.
916
June 2004 Volume 104 Number 6
Percentages of Youth Eating According to Dietary Guidelines by Weight-Control Behaviors Girls using unhealthful weight-control behaviors were less likely to meet dietary recommendations for most of the micronutrients examined (calcium, iron, vitamins C and B-6, folate, and zinc) than girls using no weightcontrol behaviors or only healthful behaviors. In addition, girls using unhealthful weight-control behaviors were less likely to consume the recommended amounts of fruits, vegetables, and grains than girls using healthful weight-control behaviors (Table 4). Only one fourth (25.7%) of girls using unhealthful weight-control behaviors consumed 1,300 mg/day or more of calcium, and only one third (33.4%) met dietary recommendations for iron intake. Analyses adjusting for race/ethnicity, SES, and school level revealed similar patterns (data not shown). Compared with girls, among boys there were fewer differences across weight-control behaviors, and differences were not consistent in their direction (Table 5). For example, boys using unhealthful weight-control behaviors were less likely to meet dietary recommendations for iron but were more likely to meet recommendations for fruit, vegetable, and saturated fat intakes than boys not engaging in weight-control behaviors. Among the boys, in analyses adjusting for race/ethnicity, SES, and school level, patterns were similar to those shown in Table 5 except for fruit, iron, and total fat intakes. For fruit and iron intake, differences across weight-control behaviors were no longer statistically significant. But, significantly more boys using unhealthful
Table 3. Dietary intake (mean/day) by types of weight-control behaviors used in the past year in adolescent boys Weight-Control Behaviors None (Nⴝ610) Energy (kcal) Fruits (servings) Vegetables (servings not including french fries) Vegetables (servings including french fries) Grains (servings) Sweet/salty snack foods (servings) Total fat (% of energy) Saturated fat (% of energy) Protein (% of energy) Carbohydrate (% of energy) Calcium (mg) Iron (mg) Vitamin A (IU) Vitamin C (mg) Vitamin B-6 (mg) Folate (g) Zinc (mg)
Healthful only (Nⴝ798)
Unhealthful (Nⴝ650)
4™™™™™™™™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™™™3 2,301⫾72 2,238⫾68 2,289⫾69 2.12⫾0.09x 2.32⫾0.08x,y 2.55⫾0.08y 1.61⫾0.08 1.64⫾0.07 1.76⫾0.08 1.80⫾0.09 1.80⫾0.08 1.95⫾0.08 6.3⫾0.2 6.3⫾0.2 6.3⫾0.2 3.3⫾0.2x 2.9⫾0.2y 3.1⫾0.2x,y 31.0⫾0.2x 30.3⫾0.2y 30.0⫾0.2y 11.3⫾0.1x 10.7⫾0.1y 10.6⫾0.1y 14.6⫾0.1 14.9⫾0.1 14.6⫾0.1 55.4⫾0.3 55.9⫾0.2 56.3⫾0.3 1,222⫾41 1,207⫾38 1,148⫾39 15.8⫾0.6 16.0⫾0.5 15.6⫾0.6 8,659⫾334 9,041⫾303 8,871⫾318 152.7⫾5.1 155.3⫾4.6 162.9⫾4.9 1.85⫾0.05 1.87⫾0.05 1.81⫾0.05 332.8⫾10.9 338.2⫾10.0 325.0⫾10.4 12.7⫾0.4 12.8⫾0.4 12.3⫾0.4
P valuea .566 .002 .401 .114 .725 .006 .003 ⬍.001 .059 .054 .107 .253 .361 .177 .349 .230 .392
a
P values are for transformed variables (except for macronutrients as % of energy and calcium intake). SE⫽standard error. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b
weight-control behaviors met dietary recommendations for total fat intake than boys reporting no weight-control behaviors (P⫽.027).
Interactions with Ethnicity/Race and Weight Status In general, ethnicity/race did not confound the relationship between dietary intake and weight-control behav-
Table 4. Percentage of adolescent girls meeting dietary recommendationsa by types of weight-control behaviors Weight-Control Behaviors
Fruits Vegetables (not including french fries) Vegetables (including french fries) Grains Total fat Saturated fat Calcium Iron Vitamin A Vitamin C Vitamin B-6 Folate Zinc
Dietary recommendation
None (Nⴝ253)
Healthful only (Nⴝ622)
Unhealthful (Nⴝ1,168)
ⱖ2 servings/day ⱖ3 servings/dayc ⱖ3 servings/day ⱖ6 servings/day ⱕ30% of energy intake ⱕ10% of energy intake ⱖ1,300 mg/day ⱖ15 mg/day ⱖ2,330 IU/day ⱖ65 mg/day ⱖ1.2 mg/day ⱖ400 g/day ⱖ9 mg/day
4™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™™3 46.8⫾3.5x,y 50.5⫾2.5x 41.2⫾2.1y x,y x 16.2⫾2.5 20.1⫾1.8 14.0⫾1.5y 17.2⫾2.6x,y 22.8⫾1.9y 16.2⫾1.6x x x 45.4⫾3.3 40.2⫾2.4 34.3⫾2.0y 47.2⫾3.6 48.9⫾2.6 51.5⫾2.2 31.6⫾3.7x 40.4⫾2.8y 45.0⫾2.5y 39.6⫾3.2x 33.6⫾2.4x 25.7⫾2.1y x x 45.7⫾3.3 42.5⫾2.3 33.4⫾1.9y 92.6⫾1.9 93.1⫾1.3 89.9⫾1.0 83.9⫾2.6x 83.1⫾1.7x,y 77.9⫾1.3y 76.4⫾3.2x 72.7⫾2.2x 63.0⫾1.8y x x 29.8⫾3.0 30.1⫾2.1 22.6⫾1.8y 66.9⫾3.3x 63.7⫾2.3x 52.4⫾1.8y
P value ⬍.001 .003 .001 .001 .356 ⬍.001 ⬍.001 .001 .049 .012 ⬍.001 ⬍.001 ⬍.001
a
Dietary recommendations are Healthy People 2010 Nutrition Objectives for fruits and vegetables and Dietary Reference Intakes for nutrients. SE⫽standard error. The recommended amount does not exclude french fries; therefore, comparisons should be made cautiously. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b c
Journal of THE AMERICAN DIETETIC ASSOCIATION
917
Table 5. Percentage of adolescent boys meeting dietary recommendationsa by types of weight-control behaviors
Fruits Vegetables (not including french fries) Vegetables (including french fries) Grains Total fat Saturated fat Calcium Iron Vitamin A Vitamin C Vitamin B-6 Folate Zinc
Weight-Control Behaviors
Dietary recommendation
None
ⱖ2 servings/day ⱖ3 servings/dayc ⱖ3 servings/day ⱖ6 servings/day ⱕ30% of energy intake ⱕ10% of energy intake ⱖ1,300 mg/day ⱖ11 mg/day ⱖ3,000 IU/day ⱖ75 mg/day ⱖ1.3 mg/day ⱖ400 g/day ⱖ11 mg/day
4™™™™™™™™™™™™™™mean⫾SE b™™™™™™™™™™™™3 41.8⫾2.6x 44.6⫾2.3x,y 48.9⫾2.5y x x,y 11.4⫾1.8 12.9⫾1.6 16.6⫾1.7y 13.9⫾1.9x 15.1⫾1.7x,y 18.9⫾1.8y 45.6⫾2.7 42.8⫾2.4 40.0⫾2.5 41.2⫾2.0 45.2⫾1.8 47.8⫾2.0 28.0⫾2.3x 37.2⫾2.0y 40.7⫾2.2y 42.0⫾2.7 40.9⫾2.5 38.8⫾2.6 67.1⫾2.8x 67.1⫾2.6x 59.9⫾2.7y 87.8⫾1.7 88.8⫾1.5 85.8⫾1.6 77.2⫾1.8 78.4⫾1.6 74.9⫾1.7 68.3⫾2.3 69.7⫾2.1 64.4⫾2.2 29.9⫾2.3 32.8⫾0.2 30.1⫾2.2 54.0⫾2.9 52.8⫾2.7 47.5⫾2.8
Healthful only
Unhealthful
P value .045 .020 .035 .151 .072 ⬍.001 .525 .007 .229 .312 .101 .419 .053
a
Dietary recommendations are Healthy People 2010 Nutrition Objectives for fruits and vegetables and Dietary Reference Intakes for nutrients. SE⫽standard error. c The recommended amount does not exclude french fries; therefore, comparisons should be made cautiously. x,y Different superscripts indicate statistically significant differences using Bonferroni adjustments. b
iors. The only exception was that African-American girls using healthful weight-control behaviors were more likely to meet dietary recommendations for vegetables (30.8%) than white girls using healthful weight-control behaviors (14.2%) (overall P⫽.007 for 8 df test, post hoc P⫽.009 for 2 df test). In contrast, vegetable intake did not differ between other African-American and white girls. Although the P values are small, inferences must be made cautiously because many outcomes were tested for interactions with ethnicity/race, and the post hoc tests were not preplanned, but rather were data driven. Overweight status did not confound any of the relationships between dietary intake and weight-control behavior in either girls or boys. DISCUSSION The current study explored associations between different types of weight-control behaviors and dietary intake among adolescent girls and boys. Among adolescent girls, the patterns of association between unhealthful weight behaviors and dietary intake were of greater concern than among the boys. Girls engaging in unhealthful weight-control behaviors had lower intakes of fruits, vegetables, grains, and all of the micronutrients assessed in this study than girls reporting only healthful weightcontrol behaviors. As compared with girls not reporting any weight-control behaviors, girls engaging in unhealthful weight-control behaviors had lower intakes of most of the micronutrients that were examined. Of particular concern were the low intakes of calcium and iron in girls using unhealthful weight-control behaviors, because intake of these nutrients tends to be less than dietary recommendations for all adolescent girls, regardless of their dieting status (33). In contrast to the girls, boys reporting weight-control
918
June 2004 Volume 104 Number 6
behaviors (healthful and unhealthful) did not differ greatly from boys not reporting any weight-control behaviors. Furthermore, the few differences that were found in the boys suggested an improved dietary intake among boys trying to control their weight. Sex differences in patterns of association between unhealthful weight-control behaviors and dietary intake found in the current study suggest that the use of unhealthful weight-control behaviors represents a different set of behaviors (and possibly associated attitudes) for boys than for girls. Further research is needed to better understand motivations and behaviors of adolescent boys reporting the use of healthful and unhealthful weight-control behaviors. Findings from the current study confirm previous findings regarding sex differences and the importance of distinguishing between healthful and unhealthful weightcontrol behaviors. In a study using data from the Youth Risk Behavior Survey, extremely unhealthful weightcontrol behaviors (defined as vomiting or diet pill use) were associated with decreased fruit and vegetable intake in girls. In contrast, moderate weight-control behaviors (defined as all other methods) were associated with increased fruit and vegetable intake and decreased highfat food intake in both girls and boys (15). The high prevalence of girls using unhealthful weightcontrol behaviors (64% of the girls) and the strong associations between these behaviors and inadequate dietary intake point to a need for education about healthful strategies for adolescent girls who are interested in weight management. It may be helpful to deter girls from “dieting,” which tends to be short-term, and encourage adoption of more healthful behaviors that can be integrated into one’s lifestyle, such as substituting fruits and vegetables for higher-fat snack foods. Findings from ethnographic research done by Nichter and colleagues (34) on
adolescent girls suggests differences between “dieting” and “watching what I eat.” “Dieting” tended to be associated with poorer dietary intake, whereas “watching” was actually associated with higher calcium intake. It may be important to teach adolescents how to “watch” what they eat, to ensure that if energy intake is reduced, nutrient intake remains stable and adequate. That said, it is important to note that even the behaviors classified in this paper as “healthful” may turn into unhealthful behaviors if done excessively. Thus, any messages promoting the use of healthful weight-control behaviors need to be given with caution. Strengths of the present study include the comprehensive assessments of weight-control behaviors and dietary intake. Most large population-based studies of youth address a broader range of adolescent health concerns and thus have only included a few questions about weightcontrol behaviors and dietary intake (15,16,21). In addition, although the study population was drawn from only one Midwestern state, its large and diverse nature in terms of ethnicity and socioeconomic backgrounds provides some confidence in making extrapolations to a broader sector of the adolescent population. However, study limitations also need to be considered. Although the time periods referred to in the questions on weightcontrol behaviors and dietary intake were similar (previous year), we cannot be sure that the behaviors were occurring simultaneously and that dietary intake was poorer during the exact time period that girls were engaging in unhealthful weight-control behaviors. Furthermore, we cannot assert that the relationship was causal and that the use of unhealthful weight-control behaviors was leading to poorer dietary intake. CONCLUSIONS Findings from the present study provide justification for interventions aimed at preventing unhealthful weightcontrol behaviors, particularly among adolescent girls. Interventions aimed at preventing unhealthful weightcontrol behaviors should include both educational and environmental components. Educational messages for both sexes can address the reasons why adolescents engage in unhealthful weight-control behaviors, the potential dangers and futility of unhealthful dieting, knowledge and skills for the use of healthful weightcontrol behaviors (when appropriate), and skills for resisting potentially harmful social influences to engage in unhealthful weight-control behaviors (35-37). Interventions should also directly address the social environments of the adolescent, including proximal factors such as weight-related norms and behaviors of family members and peers, and more distal factors, such as messages in teen magazines and other media regarding weight-control strategies, because all of these social sources influence whether and how an adolescent will try to control weight (38,39). Dietitians can play a major role in the prevention of unhealthful weight-control behaviors through their clinical work with adolescents and their families, outreach to schools and community centers, and advocacy work such as writing teen-friendly magazine articles on nutrition and weight control. Clearly the messages should focus on healthful methods to control weight and yet maintain
adequate nutrient intake. Dietitians are in a unique position in that they also have the training and skills to help adolescents at risk for obesity incorporate healthful weight-control behaviors into their lifestyle, whether for weight maintenance or gradual weight loss. References 1. Story M, French S, Resnick M, Blum R. Ethnic/racial and socioeconomic differences in dieting behaviors and body image perceptions in adolescents. Int J Eat Disord. 1995;18:173-179. 2. Serdula MK, Collins E, Williamson DF, Anda RF, Pamuk E, Byers TE. Weight control practices of U.S. adolescents and adults. Ann Intern Med. 1993;119: 667-671. 3. Neumark-Sztainer D, Hannan PJ. Weight-related behaviors among adolescent girls and boys: Results from a national survey. Arch Pediatr Adolesc Med. 2000;154:569-577. 4. Neumark-Sztainer D, Story M, Hannan PJ, Perry CL, Irving LM. Weight-related concerns and behaviors among overweight and non-overweight adolescents: Implications for preventing weight-related disorders. Arch Pediatr Adolesc Med. 2002;156:171-178. 5. Neumark-Sztainer D, Croll J, Story M, Hannan PJ, French S, Perry C. Ethnic/racial differences in weight-related concerns and behaviors among adolescent girls and boys: Findings from Project EAT. J Psychosom Res. 2002;53:963-974. 6. Spear BA. Adolescent nutrition: General, In: Worthington-Roberts BS, Williams SR, eds. Nutrition Throughout the Life Cycle. 4th ed. Boston: McGrawHill Higher Education; 2000:262-287. 7. Gong EJ, Spear BA. Adolescent growth and development: Implications for nutritional needs. J Nutr Educ. 1988;20:273-279. 8. Lytle LA. Nutritional issues for adolescents. J Am Diet Assoc. 2002;102(3 Suppl):S8-S12. 9. Spear BA. Adolescent growth and development. J Am Diet Assoc. 2002;102(3 Suppl):S23-S29. 10. Optimal calcium intake. NIH Consensus Statement Online 1994. NIH Consensus Development Conference, June 6-8, 1994; 12(4):1-31. Available at: http:// odp.od.nih.gov/Consensus/cons/097/097_statement. htm. Accessed September 25, 2003. 11. Centers for Disease Control and Prevention. Recommendations to prevent and control iron deficiency in the United States. Mor Mortal Wkly Rep CDC Surveill Summ. 1998;Apr 3:1-29. 12. Macdonald LA, Wearring GA, Moase O. Factors affecting the dietary quality of adolescent girls. J Am Diet Assoc. 1983;82:260-263. 13. Gibbons KL, Wertheim EH, Paxton SJ, Petrovich J, Szmukler GI. Nutrient intake of adolescents and its relationship to desire for thinness, weight loss behaviours, and bulimic tendencies. Aust J Nutr Diet. 1995; 52:69-74. 14. Barber SA, Bull NL. Food and nutrient intakes by British women aged 15-25 years, with particular reference to dieting habits and iron intakes. Ecol Food Nutr. 1985;16:161-169. 15. Story M, Neumark-Sztainer D, Sherwood N, Stang J, Murray D. Dieting status and its relationship to eat-
Journal of THE AMERICAN DIETETIC ASSOCIATION
919
16.
17. 18. 19.
20.
21.
22.
23.
24.
25.
26.
27.
920
ing and physical activity behaviors in a representative sample of U.S. adolescents. J Am Diet Assoc. 1998;98:1127-1135, 1255. Neumark-Sztainer D, Story M, Dixon LB, Resnick M, Blum R. Correlates of inadequate consumption of dairy products among adolescents. J Nutr Educ. 1997;29:12-20. Crawley H, Shergill-Bonner R. The nutrient and food intakes of 16-17 year old female dieters in the UK. J Hum Nutr. 1995;8:25-34. Barr SI. Dieting attitudes and behavior in urban high school students: Implications for calcium intake. J Adolesc Health. 1995;16:458-464. Webster BL, Barr SI. Calcium intakes of adolescent female gymnasts and speed skaters: Lack of association with dieting behavior. Int J Sport Nutr. 1995;5: 2-12. Mulvihill CB, Davies GJ, Rogers PJ. Dietary restraint in relation to nutrient intake, physical activity and iron status in adolescent females. J Hum Nutr Diet. 2002;15:19-31. Middleman AB, Vazquez I, Durant RH. Eating patterns, physical activity, and attempts to change weight among adolescents. J Adolesc Health. 1998; 22:37-42. Brugman E, Meulmeester JF, Spee-van der Wekke A, Beuker RJ, Zaadstra BM, Radder JJ, Verloove-Vanhorick PS. Dieting, weight and health in adolescents in The Netherlands. Int J Obesity. 1997;21:54-60. Neumark-Sztainer D, Story M, Resnick MD, Blum RW. Correlates of inadequate fruit and vegetable consumption among adolescents. Prev Med. 1996;25: 497-505. Neumark-Sztainer D, Story M, Hannan P, Moe J. Overweight status and eating patterns among adolescents: Where do youth stand in comparison to the Healthy People 2010 Objectives? Am J Public Health. 2002;92:844-851. Neumark-Sztainer D, Wall M, Story M, Perry CL. Correlates of unhealthful weight-control behaviors among adolescent girls and boys: Implications for the primary prevention of disordered eating. Health Psychol. 2003;22:88-98. Neumark-Sztainer D, Hannan PJ, Story M, Croll J, Perry C. Family meal patterns: Associations with sociodemographic characteristics and improved dietary intake among adolescents. J Am Diet Assoc. 2003;103:317-322. Rockett HRH, Breitenbach MA, Frazier AL, Witschi J, Wolf AM, Field AE, Colditz GA. Validation of a youth/adolescent food frequency questionnaire. Prev Med. 1997;26:808-816.
June 2004 Volume 104 Number 6
28. Rockett HR, Wolf AM, Colditz GA. Development and reproducibility of a food frequency questionnaire to assess diets of older children and adolescents. J Am Diet Assoc. 1995;95:336-340. 29. Healthy People 2010. Washington, DC: US Dept of Health and Human Services; 2000. 30. Yates AA, Schlicker SA, Suitor CW. Dietary Reference Intakes: The new basis for recommendations for calcium and related nutrients, B vitamins, and choline. J Am Diet Assoc. 1998;98:699-706. 31. Food and Nutrition Board, Institute of Medicine, National Academy of Sciences. New dietary reference intakes: Recommended levels for individual intakes 1997 and 1998, recommended daily allowances, revised 1989. Nutr Today. 1998;33:257-260. 32. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board, Institute of Medicine. New dietary reference intakes for calcium and related nutrients. Nutrition Research Newsletter. 1997;16:85-87. 33. Alaimo K, McDowell MA, Briefel RR, Bischof AM, Caughman CR, Loria CM, Johnson CL. Dietary intake of vitamins, minerals, and fiber of persons ages 2 months and over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988-91. Washington, DC: National Center for Health Statistics; 1994. Advance Data from Vital and Health Statistics. 258:1-28. 34. Nichter M, Ritenbaugh C, Nichter M, Vuckovic N, Aickin M. Dieting and “watching” behaviors among adolescent females: Report of a multimethod study. J Adolesc Health. 1995;17:153-162. 35. Neumark-Sztainer D, Story M, Hannan PJ, Rex J. New Moves: A school-based obesity prevention program for adolescent girls. Prev Med. 2003;37:41-51. 36. Littleton HL, Ollendick T. Negative body image and disordered eating behavior in children and adolescents: What places youth at risk and how can these problems be prevented? Clin Child Fam Psychol Rev. 2003;6:51-66. 37. Paxton SJ, Wertheim EH, Pilawski A, Durkin S, Holt T. Evaluations of dieting prevention messages by adolescent girls. Prev Med. 2002;35:474-491. 38. Martinez-Gonzalez MA, Gual P, Lahortiga F, Alonso Y, de Irala-Estevez J, Cervera S. Parental factors, mass media influences, and the onset of eating disorders in a prospective population-based cohort. Pediatrics. 2003;111:315-320. 39. Utter J, Neumark-Sztainer D, Wall M, Story M. Reading magazine articles about dieting and associated weight-control behaviors among adolescents. J Adolesc Health. 2003;32:78-82.