Abstract
Objective
In this research study, we examine how both overweight and nonoverweight adolescent girls and boys fared from 1999 to 2010 in terms of disordered eating behaviors and psychosocial well-being.
Method
A repeated cross-sectional design was used. Participants were recruited from public schools in 1999 (n = 3072, mean age = 14.6 ± 1.8) and 2010 (n = 2793, mean age = 14.4 ± 2.0). Secular trends were examined by weight status and gender using inverse probability weighting to control for changes in socio-demographics.
Results
In general, the prevalence of disordered eating behaviors and markers of psychosocial well-being among overweight girls and boys remained the same from 1999 to 2010. In contrast, among nonoverweight girls, chronic dieting, unhealthy weight control behaviors, and extreme weight control behaviors decreased, and body satisfaction improved during this time period. Further, among non-overweight boys, the prevalence of unhealthy and extreme weight control behaviors decreased, as did mean depression scores.
Discussion
Overall, findings indicate a strong need to ensure that messages about the dangers of disordered eating behaviors are reaching overweight youth. Obesity prevention interventions should not overlook the comorbid nature of obesity, disordered eating and poor psychosocial health; prevention programming should address shared risk factors, including dieting, media use, body dissatisfaction, and weight-related teasing.
Keywords: obesity, secular trends, weight control, psychosocial well-being, disordered eating, dieting, adolescents
Introduction
Disordered eating behaviors1,2 and markers of poor psychosocial health, including depression, poor self-esteem, and low body satisfaction are prevalent among adolescents.1 Further, evidence suggests that the prevalence of disordered eating and markers of poor psychosocial health are higher among overweight adolescents as compared with their nonoverweight peers.2–5 Two recently published studies, including one conducted on the sample analyzed for this study, have explored how adolescents’ use of disordered eating behaviors changed over time (i.e., ecular trends).1,2 Both studies found that the overall prevalence of disordered eating behaviors decreased over observed time periods (trends examined from 1999 to 2009 in one study and from 1999 to 2010 in the other) among adolescent girls, with less consistent trends in boys. While the prevalence of adolescents engaging in disordered eating behaviors remains high, findings from these two studies paint an encouraging picture, suggesting a decrease in the overall prevalence of these behaviors among adolescents over time.
However, these studies did not examine secular trends by weight status. Given the strong public health messages regarding obesity over recent years, it is crucial to determine if and how disordered eating behaviors and psychosocial well-being have changed among overweight youth. In this research study, we examine how both overweight and nonoverweight adolescent girls and boys fared from 1999 to 2010 in terms of disordered eating behaviors and psychosocial well-being.
Method
Study Design and Population
A repeated cross-sectional study design was used to compare weight-related variables among adolescents using data collected from in 1999 (Project EAT-I; n = 3,072)5–7 and from a new cohort of adolescents in 2010(EAT 2010; n = 2,793).2,8,9 Study procedures were approved by the University of Minnesota's Institutional Review Board Human Subjects Committee and by the research boards of the participating school districts. Additional details on data collection have been previously described.6,8
Measures
Survey development was guided by a theoretical framework,10,11 expert review, qualitative input from adolescents,12 and pilot testing. Test-retest reliability was assessed in diverse adolescents at EAT-I (n = 161) and at EAT 2010 (n = 129); psychometric properties from 2010 are reported below. The subsample utilized for test-retest at EAT 2010 was 70% non-Hispanic white and equally divided on gender.
Disordered Eating Behaviors
Chronic dieting was assessed by asking participants: “How often have you gone on a diet during the last year?”. Responses were dichotomized to distinguish chronic dieters indicating, “I am always dieting” from others selecting one of the other response options (Test-retest agreement = 97%).
Unhealthy and extreme weight control behaviors were assessed by asking participants: “Have you done any of the following things to lose weight or keep from gaining weight during the past year?” Endorsement of behaviors (any vs. none) were divided into two categories, unhealthy (fasted, ate very little food, used a food substitute, skipped meals, smoked more cigarettes) and extreme (took diet pills, made myself vomit, used laxatives, used diuretics). (Test-retest agreement = 85% for unhealthy behaviors and 95% for extreme behaviors).
Binge eating was assessed with up to two yes/no questions: (1) “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)” (Test-retest agreement = 90%) and, if yes, “During the times when you ate this way, did you feel you could not stop eating or control what or how much you were eating?” (Test-retest agreement = 75%).
Psychosocial Well-Being
Depressive symptoms were assessed using a six-item scale developed by Kandel and Davies.13 Higher scale values were indicative of more severe depressed mood (Cronbach's α = 0.83; Test–retest r = 0.79).
Self-esteem was measured with six items from the Rosenberg Self-esteem Scale.14 Responses were summed with higher scores indicative of greater self-esteem (Score range: 6–24; Cronbach's α = 0.77; Test–retest r = 0.69).
Body satisfaction was assessed with a modified version of the Body Shape Satisfaction Scale.15 Responses were summed with higher scores indicative of greater body satisfaction (Score range: 10–50; Cronbach's α = 0.93; Test–retest r = 0.67).
Weight concerns were assessed by asking adolescents to indicate their agreement with two statements: (a) “I think a lot about being thinner” and (b) “I am worried about gaining weight”. Item responses were summed with higher scores indicative of greater weight concern (Score range: 2–8; Cronbach's α = 0.83; Test–retest r = 0.77).
Weight Status
BMI values were calculated from measured heights and weights obtained by trained research staff using standardized protocols. Sex- and age-specific cutoff points used to classify respondents as not overweight (BMI < 85th percentile) or overweight (BMI ≥ 85th percentile).16
Statistical Analysis
To test for secular changes in disordered eating and psychological well-being outcomes, we first controlled for demographic shifts in the study population from 1999 to 2010 but utilizing inverse probability weighting.17 Details on this approach have been previously published.2 Proper control of the demographic shift using the weights was achieved as evidenced by the nonsignificant differences in the weighted 1999 sample compared to the 2010 sample of both nonoverweight and overweight youth (Table 1).
TABLE 1.
Socio-demographic characteristics of 1999 and 2010 nonoverweight (<85th percentile) and overweight (>85th percentile) adolescent study participants
| NonOverweight (<85th percentile) |
Overweight (>85th percentile) |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1999 (n = 1784) | 1999 Weighted (n = 1784) | 2010 (n = 1646) | 1999 (n = 896) | 1999 Weighted (n = 896) | 2010 (n = 1101) | |||||||||
| N | % | N | % | N | % | p valuesa | N | % | N | % | N | % | p valuesb | |
| Gender | 0.942 | 0.979 | ||||||||||||
| Male | 887 | 49.7 | 812 | 45.5 | 747 | 45.4 | 423 | 47.2 | 436 | 48.7 | 535 | 48.6 | ||
| Female | 897 | 50.3 | 972 | 54.5 | 899 | 54.6 | 473 | 52.8 | 460 | 51.3 | 566 | 51.4 | ||
| Ethnicity/race | 0.999c | 0.999c | ||||||||||||
| White | 626 | 35.1 | 400 | 22.4 | 366 | 22.2 | 298 | 33.3 | 128 | 14.3 | 155 | 14.1 | ||
| Black | 380 | 21.3 | 501 | 28.1 | 465 | 28.3 | 216 | 24.1 | 260 | 29.0 | 325 | 29.5 | ||
| Asian | 480 | 26.9 | 369 | 20.7 | 339 | 20.6 | 188 | 21.0 | 174 | 19.4 | 209 | 19.0 | ||
| Hispanic | 121 | 6.8 | 271 | 15.2 | 253 | 15.4 | 86 | 9.6 | 175 | 19.5 | 215 | 19.5 | ||
| Native American | 65 | 3.6 | 39 | 2.2 | 38 | 2.3 | 53 | 5.9 | 49 | 5.5 | 59 | 5.4 | ||
| Mixed/Other | 86 | 4.8 | 166 | 9.3 | 180 | 10.9 | 55 | 4.7 | 86 | 9.6 | 134 | 12.2 | ||
| Not reported | 26 | 1.5 | 36 | 2.0 | 5 | 0.3 | 13 | 1.5 | 25 | 2.8 | 4 | 0.4 | ||
| Socio-economic status | 0.999 | 0.999 | ||||||||||||
| Low | 373 | 20.9 | 633 | 35.5 | 586 | 35.6 | 212 | 23.7 | 375 | 41.9 | 472 | 42.9 | ||
| Low-middle | 326 | 18.3 | 373 | 20.9 | 346 | 21.0 | 194 | 21.7 | 195 | 21.8 | 235 | 21.3 | ||
| Middle | 441 | 24.7 | 308 | 17.3 | 282 | 17.1 | 233 | 26.0 | 149 | 16.7 | 182 | 16.5 | ||
| High-middle | 316 | 17.7 | 251 | 14.1 | 229 | 13.9 | 147 | 16.4 | 95 | 10.7 | 116 | 10.5 | ||
| High | 241 | 13.5 | 163 | 9.2 | 151 | 9.2 | 76 | 8.5 | 41 | 4.6 | 49 | 4.5 | ||
| Not reported | 87 | 4.9 | 56 | 3.2 | 52 | 3.2 | 34 | 3.8 | 40 | 4.4 | 47 | 4.3 | ||
| N | Mean (SD) | N | Mean (SD) | N | Mean (SD) | p values | N | Mean (SD) | N | Mean (SD) | N | Mean (SD) | p values | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age in years | 1784 | 14.7 (1.8) | 1784 | 14.6 (1.8) | 1646 | 14.5 (2.0) | 0.455 | 896 | 14.5 (1.8) | 896 | 14.3 (1.8) | 1101 | 14.3 (2.0) | 0.718 |
p-values are presented for differences between the weighted 1999 nonoverweight sample and the 2010 nonoverweight sample; based on χ2 tests for gender, ethnicity/race, and socio-economic status and t-tests for age.
p-values are presented for differences between the weighted 1999 overweight sample and the 2010 overweight sample (similar tests as above).
Test for race combines the “mixed/other” and “not reported” groups.
Tests for secular trends in continuous and dichotomous outcome variables were conducted using two sample t-tests and χ2 tests, respectively, with the inverse probability weights incorporated for the 1999 sample. All analyses were stratified by overweight status and gender. Reported p-values were not adjusted for multiple testing. All analyses were performed in SAS 9.2.
Results
Overweight girls
Small decreases in the percentage of overweight girls engaging in disordered eating behaviors were observed between 1999 and 2010; however, none of these observed decreases reached statistical significance (Table 2). Further, no statistically significant secular changes in the level of weight concern, body satisfaction, depressive symptoms, or self-esteem reported by overweight girls were observed.
TABLE 2.
Trends in disordered eating behaviors and related psychosocial health/attitudes among nonoverweight and overweight adolescents in Minneapolis/St. Paul from 1999 to 2010 (percents and means)
| Nonoverweight girls (BMI <85th percentile) | Overweight girls (BMI ≥ 85th percentile) | Nonoverweight boys (BMI <85th percentile) | Overweight boys (BMI ≥ 85th percentile) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1999a | 2010 | ESb | p values | 1999a | 2010 | ESb | p values | 1999a | 2010 | ESb | p values | 1999a | 2010 | ESb | p values | |
| n= 972c | n=899c | n=460c | n=566c | n=812c | n=747c | n=436c | n=535c | |||||||||
| Disordered eating behaviors (%) | %diff | %diff | %diff | %diff | ||||||||||||
| Chronic dieting | 7.4 | 2.7 | –4.7 | <.01 | 12.9 | 10.1 | –2.8 | 0.15 | 3.3 | 1.9 | –1.4 | 0.09 | 10.9 | 6.9 | –4.0 | 0.03 |
| Less extreme weight control behaviors | 47.7 | 38.7 | –9.0 | <.01 | 72.8 | 68.1 | –4.7 | 0.10 | 29.0 | 23.9 | –5.1 | 0.02 | 58.3 | 57.9 | –0.4 | 0.90 |
| Extreme weight Control behaviors | 8.9 | 4.0 | –4.9 | <.01 | 13.9 | 11.3 | –2.6 | 0.21 | 5.1 | 3.0 | –2.1 | 0.04 | 5.8 | 5.4 | –0.4 | 0.81 |
| Binge eating With loss of control | 8.9 | 7.8 | –1.1 | 0.40 | 12.3 | 12.1 | –0.2 | 0.94 | 3.1 | 4.2 | 1.1 | 0.27 | 7.6 | 9.1 | 1.5 | 0.39 |
| Psychosocial well-being (mean) | Mean diff | Mean diff | Mean diff | Mean diff | ||||||||||||
| Depressive symptoms | 18.1 | 17.9 | –0.04 | 0.29 | 18.0 | 18.5 | 0.10 | 0.10 | 15.9 | 15.4 | –0.10 | 0.04 | 16.2 | 16.3 | 0.02 | 0.74 |
| Self-esteem | 17.6 | 17.7 | 0.03 | 0.56 | 17.3 | 17.2 | –0.03 | 0.76 | 18.5 | 18.8 | 0.08 | 0.22 | 18.0 | 17.6 | –0.11 | 0.06 |
| Body satisfaction | 34.7 | 35.7 | 0.10 | 0.03 | 30.3 | 29.4 | –0.09 | 0.16 | 38.7 | 37.4 | –0.13 | <0.01 | 32.7 | 30.9 | –0.18 | 0.01 |
| Weight concerns | 2.5 | 2.2 | –0.30 | <.01 | 3.1 | 3.0 | –0.10 | 0.07 | 1.9 | 1.7 | –0.20 | <.001 | 2.8 | 2.6 | –0.20 | 0.01 |
| BMI (SD=3.0 for non overweight, 4.7 for overweight) | 20.5 | 20.3 | –0.07 | 0.03 | 29.1 | 29.4 | 0.06 | 0.39 | 20.3 | 20.0 | –0.10 | <0.01 | 28.6 | 28.8 | 0.04 | 0.59 |
The 1999 sample was weighted to allow for an examination of secular trends in weight-related outcomes independent of demographic shifts in the population.
ES = Effect size for dichotomous outcomes is difference in proportions and for continuous outcomes is standardized mean difference (i.e., mean diff is scaled by the full sample standard deviation). For BMI a different standard deviation was used to standardize the nonoverweight and overweight since the groups differed greatly by BMI by design.
Numbers may vary slightly due to missing values for specific variables.
Nonoverweight girls
The percentage of nonoverweight adolescent girls engaging in chronic dieting, unhealthy and extreme weight control behaviors, decreased significantly from 1999 to 2010 (Table 2). A decrease in weight concerns was seen among nonoverweight girls (ES = −3.0, p < 0.01). Body satisfaction improved among nonoverweight girls (ES = 0.10, p = 0.03).
Overweight boys
The percentage of overweight boys who reported chronic dieting decreased significantly from 1999 to 2010 (ES = −4.0, p = 0.03) (Table 2). A small, but statistically significant, decrease in weight concerns was seen among overweight boys (ES = −0.20, p = 0.01). Finally, body satisfaction decreased among overweight boys (ES = −0.18, p = 0.01).
Nonoverweight boys
The percentage of nonoverweight boys engaging in unhealthy and extreme weight control behaviors decreased significantly from 1999 to 2010 (Table 2). A small, but statistically significant, decrease in depressive symptoms among nonoverweight adolescent boys was seen (ES = −0.10, p = 0.04). Body satisfaction decreased slightly from 1999 to 2010 (ES = −0.13, p < 0.01).
Discussion
This study examined secular trends in disordered eating behaviors and psychosocial well-being from 1999 to 2010 among adolescents by weight status. This study's main finding is that although significant secular decreases in the percentage of nonoverweight girls engaging in disordered eating behaviors were observed between 1999 and 2010, parallel significant decreases were not observed among overweight girls. Similarly, more significant secular decreases in use of disordered eating behaviors were observed among nonoverweight boys as compared with overweight boys; however, overall patterns were not as consistent among boys. Findings suggest a strong need to ensure that messages about the dangers and ineffectiveness of disordered eating behaviors are reaching overweight adolescents.
Several noteworthy secular trends were observed among nonoverweight adolescents from 1999 to 2010, including a decrease in disordered eating and improvement in psychosocial well-being. Specifically, chronic dieting and unhealthy and extreme weight control behaviors decreased and body satisfaction improved among nonoverweight girls. While these changes are encouraging, it is important to note that the percentages of girls engaging in disordered eating behaviors remains high and the prevention of these behaviors should continue to be a top public health concern. Among nonoverweight boys, there were decreases in unhealthy and extreme weight control behaviors, and mean depression scores, between 1999 and 2010. It was also encouraging that, in general, among overweight adolescents, disordered eating and psychosocial well-being did not appear to worsen from 1999 to 2010, despite the increased attention on obesity-related issues in the media during this time period. Unfortunately, the overall trends of decreased use of disordered eating behaviors and improvement in psychosocial well-being observed in nonoverweight youth were not reflected in the youth who were overweight. Further, a decrease in the mean level of body satisfaction among overweight boys was observed between 1999 and 2010.
Study strengths and limitations should be taken into account when interpreting study results. The unique repeated cross-sectional study design allowed for the study of secular trends from 1999 to 2010, a period during which much media attention was directed to weight-related topics. The large and diverse study population, measured heights and weights, and breadth of questions are additional study strengths. However, all study data were collected from participants in one urban area in Minnesota, limiting the overall generalizability of study findings.
Overall, findings indicate a need to ensure that messages about the dangers of disordered eating behaviors are reaching overweight youth. Interventions aimed at obesity prevention should not overlook the comorbid nature of obesity, disordered eating, and markers of poor psychosocial health; prevention programming should focus on addressing shared risk factors.18–20
References
- 1.Eaton DK, Kann L, Kinchen S, Shanklin S, Flint KH, Hawkins J, et al. Youth risk behavior surveillance - United States, 2011. MMWR Surveill Summ. 2012;61:1–162. [PubMed] [Google Scholar]
- 2.Neumark-Sztainer D, Wall MM, Larson N, Story M, Fulkerson JA, Eisenberg ME, et al. Secular trends in weight status and weight-related attitudes and behaviors in adolescents from 1999 to 2010. Prev Med. 2012;54:77–81. doi: 10.1016/j.ypmed.2011.10.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Loth KA, Mond J, Wall M, Neumark-Sztainer D. Weight status and emotional Well-being: longitudinal findings from project EAT. J Pediatr Psychol. 2011;36:216. doi: 10.1093/jpepsy/jsq026. [DOI] [PubMed] [Google Scholar]
- 4.Neumark-Sztainer D, Wall MM, Haines JI, Story MT, Sherwood NE, van den Berg PA. Shared risk and protective factors for overweight and disordered eating in adolescents. Am J Prev Med. 2007;33:359–369. doi: 10.1016/j.amepre.2007.07.031. [DOI] [PubMed] [Google Scholar]
- 5.Neumark-Sztainer D, Story M, Hannan PJ, Perry CL, Irving LM. Weight-related concerns and behaviors among overweight and nonoverweight adolescents: Implications for preventing weight-related disorders. Arch Pediatr Adolesc Med. 2002;156:171. doi: 10.1001/archpedi.156.2.171. [DOI] [PubMed] [Google Scholar]
- 6.Neumark-Sztainer D, Story M, Hannan PJ, Croll J. Overweight status and eating patterns among adolescents: Where do youths stand in comparison with the healthy people 2010 objectives? Am J Public Health. 2002;92:844. doi: 10.2105/ajph.92.5.844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Neumark-Sztainer D, Wall M, Guo J, Story M, Haines J, Eisenberg M. Obesity, disordered eating, and eating disorders in a longitudinal study of adolescents: How do dieters fare 5 years later? J Am Diet Assoc. 2006;106:559–568. doi: 10.1016/j.jada.2006.01.003. [DOI] [PubMed] [Google Scholar]
- 8.Larson NI, Wall MM, Story MT, Neumark-Sztainer DR. Home/family, peer, school, and neighborhood correlates of obesity in adolescents. Obesity. 2013;21:1858–1869. doi: 10.1002/oby.20360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wall MM, Larson NI, Forsyth A, Van Riper DC, Graham DJ, Story MT, et al. Patterns of obesogenic neighborhood features and adolescent weight: A comparison of statistical approaches. Am J Prev Med. 2012;42:e65–75. doi: 10.1016/j.amepre.2012.02.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bandura A. Social foundations of thought and action: A social cognitive theory. Prentice-Hall; Englewood Cliffs, NJ: 1986. [Google Scholar]
- 11.Sallis JF, Owen N, Fisher EB. Ecological models of health behavior. Health behavior and health education: Theory, research, and practice. 2008;4:465–485. [Google Scholar]
- 12.Neumark-Sztainer D, Story M, Perry C, Casey MA. Factors influencing food choices of adolescents: Findings from Focus-group discussions with adolescents. J Am Diet Assoc. 1999;99:929–37. doi: 10.1016/S0002-8223(99)00222-9. [DOI] [PubMed] [Google Scholar]
- 13.Kandel DB, Davies M. Epidemiology of depressive mood in adolescents: An empirical study. Arch Gen Psychiatry. 1982;39:1205–1212. doi: 10.1001/archpsyc.1982.04290100065011. [DOI] [PubMed] [Google Scholar]
- 14.Rosenberg M. [2014 2, 11];Society and the adolescent self-image. 1965 Available at: http://psycnet.apa.org/journals/ort/36/3/560.pdf&productCode=pa.
- 15.Pingitore R, Spring B, Garfieldt D. Gender differences in body satisfaction. Obes Res. 1997;5:402–409. doi: 10.1002/j.1550-8528.1997.tb00662.x. [DOI] [PubMed] [Google Scholar]
- 16.Kuczmarski RJ, Ogden CL, Guo SS, Grummer-Strawn LM, Flegal KM, Mei Z, et al. 2000 CDC growth charts for the united states: methods and development. Vital Health Stat 11. Data from the National Health Survey. 2002;1 [PubMed] [Google Scholar]
- 17.Robins JM, Finkelstein DM. Correcting for noncompliance and dependent censoring in an AIDS clinical trial with inverse probability of censoring weighted (IPCW) Log-rank tests. Biometrics. 2000;56:779–788. doi: 10.1111/j.0006-341x.2000.00779.x. [DOI] [PubMed] [Google Scholar]
- 18.Haines J, Neumark-Sztainer D. Prevention of obesity and eating disorders: A consideration of shared risk factors. Health Edu Res. 2006;21:770–782. doi: 10.1093/her/cyl094. [DOI] [PubMed] [Google Scholar]
- 19.Irving LM, Neumark-Sztainer D. Integrating the prevention of eating disorders and obesity: Feasible or futile? Prev Med. 2002;34:299–309. doi: 10.1006/pmed.2001.0997. [DOI] [PubMed] [Google Scholar]
- 20.Neumark-Sztainer D. Obesity and eating disorder prevention: An integrated approach? Adolesc Med. 2003;14:159–173. [PubMed] [Google Scholar]