What is the current best-practice for prevention of eating disorders among adolescent/young adults based on reduction of eating disorder risk factors (as determined by previously validated eating disorder behavior questionnaires)?
Elizabeth Parker, RD and Margery Lawrence, PhD, RD
December 6, 2017
Introduction: The purpose of this study was to determine the current best intervention program for reducing the risk of developing an eating disorder (ED), based on previously validated eating disorder behavior questionnaires. With over 10% of adolescent/young adult females developing eating disorders, prevention is increasingly important. Method: An evidence-based review was conducted looking at programs designed to reduce risk factors that have been shown to be associated with developing an eating disorder among high school and college students. The research found was limited to females, and primarily focused on those experiencing body dissatisfaction. Results: Twelve studies were reviewed, nine of which found cognitive-dissonance prevention programs to be significantly more effective than control groups. Conclusion: Cognitive dissonance-based programing (the Body Project) was found to be the most effective way to change negative beliefs and behaviors that lead to EDs among females. Further research should include studies that look at broader population bases, creating programming for males and those of the lesbian, gay, bisexual, transgender, and queer community (LGBTQ).
The prevalence of eating disorders (ED) has grown quite high, with approximately 10% (or more) of adolescent girls/young women meeting the diagnostic criteria for eating disorders based on the DSM-IV or DSM-5.1, 3, 4, 6, 7, 9, 10, 11 Additionally, roughly 50% of college-age women may have subclinical disordered eating behaviors.10 Behaviors and conditions commonly associated with disordered eating include: dietary restraint, purging (self-induced vomiting, over exercise, laxatives, diuretics, diet pills), binge-eating, self-deprecation, negative affect, difficulty regulating emotion, anxiety, depression, self-harm, thin-ideal internalization, and body dissatisfaction to name a few.
Many women who do receive treatment do not fully recover from their ED (e.g. 44% of those with bulimia nervosa do not, and that percentage is greater for those suffering from anorexia nervosa)10 which is why prevention of these disorders is so important. Early assessment and intervention is crucial for the best odds of recovery and restoration of health, especially in those under the age of 19.4 EDs generally develop in adolescence and prevention programs are more effective in early adolescence.6 On average, body dissatisfaction peaks in later adolescence but these older adolescents may be able to better comprehend the need for intervention of socially promoted “thin ideal.”6
It is concerning that only one-third of people with an ED have been asked about eating-related issues by their primary health care provider.4 Furthermore, less than one-third of people with EDs receive treatment!4, 6 Given that more people will see a primary medical provider than a specialist, primary care providers have an opportunity to assess patients for eating disorders and should be trained in how to screen for them.4 Many medical providers said they felt “ill-equipped” to screen for or treat eating disorders.4 Since schools have a unique opportunity to track and influence students over a longer period of time, many studies have looked at prevalence and prevention strategies among students.
Prevention programs are relatively new in the field of ED research, and are not widely practiced. In fact, most of the available studies performed to evaluate these programs were performed by the same pool of researchers. The purpose of this evidence-based review was to determine the current best intervention program for reducing the risk of developing an eating disorder, based on previously validated eating disorder behavior questionnaires.
Several programs were seen repeatedly among the various studies, suggesting that there was already some consensus on what programs might yield the most significant reductions in ED risk factors. The studies looked at reduction of ED risk factors, especially surrounding thin-ideal internalization, dieting practices, body dissatisfaction, depression/negative affect, and used similar self-reported scales.
Overall, the studies indicate that features of useful ED prevention programs include: cognitive-dissonance-based programing, multiple sessions (instead of a single session), interactive sessions (group, or other interaction), facilitation by professionals (as opposed to peers, teachers or other endogenous staff), use of validated assessments, and avoiding education about EDs and effects of EDs.
A Pubmed search was performed on August 21st, 2017 using the following search criteria: ((feeding and eating disorders/prevention & control*)) AND ((young adult OR adolescent) AND students) AND (risk factors AND eating disorders) AND (“last 10 years”[PDat] AND Humans[Mesh] AND English[lang])
The search yielded 35 articles, twenty of which were initially selected based on the abstract. This was further reduced to twelve by excluding any age groups that were not high school or college (undergraduate/graduate level). Since this is a relatively new topic (especially in areas of peer leaders; 2,9,12 internet-based program; 7 and female athletes 10), all study designs were initially included due to the small number of available studies. The significance was held at p= or <0.05 for all studies.
Of the resulting 35 articles, 23 were excluded due to: age of population, abstract not indicating ED prevention, or study types such as interview, reviews or meta-analyses. Twelve articles were selected for review as follows in table 1.
Of the 12 articles, 10 looked at cognitive dissonance-based (CD) prevention programs. Of those 10, 9 concluded that CD programs produced more significant reductions in ED risk factors than other prevention programs. The one exception was the study that focused specifically on athletes.10
There are previously validated assessments for mental health and eating disorders as seen in the assessment methods used in the studies (see “key to abbreviations” at the end of this paper). These assessments were used to determine if there was a reduction in negative thoughts and disordered behaviors among participants.
The four most common interventions examined in studies were in-person cognitive dissonance-based programs (CD), internet-based cognitive-dissonance based programs, the NEDA brochure, and the documentary “Dying To Be Thin.” The most effective of these was in-person cognitive dissonance-based programs.
Cognitive Dissonance-Based (CD) ED programs (The “Body Project”):
Cognitive dissonance “is based on the presumption that creating an inconsistency between a belief and a behavior will elicit a feeling of discomfort in an individual”2 and that the individual will need to change the behavior to remove the discomfort. Programs focused on CD help to promote behavioral changes because consistency between beliefs and behaviors is human nature. By challenging disordered beliefs, behaviors are challenged; and to remain consistent behaviors must change along with the belief.7
The Body Project is an in-person CD group program, typically: 4 weekly 1-hour sessions, with 7-9 participants per group, led by a trained (for 9 hours) and scripted facilitator (counselor or peer).1 CD programs had the greatest effect on reducing ED symptoms in all the studies that included the “Body Project” as an intervention. 1, 2, 3, 4, 6, 9, 11, 12 CD challenges the thin ideal by creating cognitive dissonance with written, verbal, and behavioral exercises.1 With adequate training it worked for peers to “teach” the material, but not facilitate difficult discussions.10 Group programs, like the Body Project, further increase behavior change by the extra layer of public accountability.7
The Body Project has the greatest amount of research (relative to similar programs) and has a strong evidence-base.12 All researchers who used this intervention cited efficacy trials for dissonance-based programs that showed promise. The American Psychological Association (APA) supports dissonance-based prevention programs for interventions, for creating replicable results, and creating significant results against the control groups at 2 and 3 years, and to a 60% greater effect than assessment-only control groups.10 Many of the study authors commented that more sessions may have produced greater results.2, 4, 10
Internet-based Cognitive Dissonance Based Programming (The “eBody Project”):
The e-Body Project is CD-based like the Body Project, but delivered via the internet. The program was designed to remain interactive, and contains 6 modules to be completed over 3 weeks, at 30-40 minutes/session.1 The internet has become a constant source of thin-ideal, so to have the internet challenge that ideal through this program is very timely.1
Two studies reported that many schools and other settings found it difficult to recruit clinicians (school counselors/nurses/teachers) to learn the Body Project and lead the groups.7, 9 With over 95% of adolescents having internet access,1 a larger population can be reached via the internet.
The e-Body Project showed promising results where the in-person Body Project could not be performed. The ED risk reduction results were not as strong which may be due to the fact that participants could go through this program faster, meaning that there was less time submerged in the messaging. This had less impact on reduction of thin-ideal internalization.7
“Dying To Be Thin”:
This is a 55-minute documentary (McPhee, 2000). Widely-available at no-cost, this video covers body image, pressures to be thin, eating disorders, treatment, recovery, and consequences of eating disorders.1
This is a two-page brochure (National Eating Disorders Association, 2002) covering negative and positive body image and how negative body image may lead to eating disorders. The brochure lists ten steps to positive body image.1, 11
Quality of Studies:
The studies were inconsistent in quality. Though the majority (8 of 12) were randomized control trials (RCT), the criteria and methods left many of them lacking, with five of the twelve receiving a negative (-) grade. Four were well organized and received a positive (+) rating. Most of the studies had a strong conclusion, but due to lack of generalizability (homogeneity) or sample size, most (8 of 12) were categorized as a grade II, with only one study deserving of a grade I. Overall, the grade of research is: II, neutral.
All of the studies focused solely on females, with relatively homogeneous populations. Demographically, studies were representative of female students at American universities, but not the population at large.1, 3, 4, 5, 7, 11 Additionally, all of the studies had sample sizes of less than 1000 participants, the majority having less than 200.
There was a great likelihood of selection bias in all of the studies, as participants could voluntarily opt-in to the study or opt-out at any time. Several researchers noted that the participants who stayed in the studies may have had more interest in the topic of body image. Therefore, it is difficult to draw conclusions on how effective the interventions would be with a more universal population. Furthermore, as the study assessments were self-reported surveys, there was potential for error in interpretation of questions, or in answering questions how they think the researchers want to hear.
In conclusion, based on the studies reviewed, cognitive dissonance-based prevention programs in group settings (such as the Body Project) created the greatest reduction in ED risk factors, even at 3 years-post intervention, compared to control interventions. These CD-based programs have a greater likelihood of success in settings such as schools, especially Colleges/Universities. This is because of the unique setting schools offer in which it is often easier to deliver prevention programs. Although trained counseling clinicians were most effective, several studies noted that training peer leaders or other endogenous staff (such as teachers or school nurses) led to a statistically significant reduction of ED risk factors compared to control groups.
The one study that had participants go to a medical clinic, rather than having programming done at school, had high dropout rates.4 The authors explained that this was due to the greater time and effort needed to go to an off-campus clinic. This further shows how much more effective campus settings can be for prevention programs. With over 11.5 million female college students in 2017 in the U.S.13 (meaning approximately 1.15 million of which will develop a diagnosable eating disorder, given a 10% prevalence) this is utterly important.11
Universal programs (inclusive of all genders):
It was stated in one paper that there was research showing that programs targeting high-risk participants led to greater effects than programs that were “universal.”9 However, none of the studies evaluated for this review were universal, in that they were aimed at only females with body image concerns who did not have a diagnosed eating disorder. We believe that further studies could be done to assess whether or not cognitive dissonance-based prevention programs would be effective for males or those of the LGBTQ community. Due to body image differences among genders/gender identity, there would likely have to be different content for the CD programming to make it effective.
Programs for those with active EDs:
Potential participants with diagnosable EDs were excluded from all of the studies reviewed. This leads us to believe there is a gap in the research for programs that will reduce ED risk factors in those with clinical levels of EDs. This may be because at the point of having a diagnosable ED, it is not “prevention” anymore, but rather treatment. The disheartening piece is that according to background research in two studies (Linville, Cobb, Lenee-Bluhm, et. al; and Muller and Stice) over two-thirds of people who have EDs do not get diagnosed or receive treatment.4,6 Further research in programming that can help those with active EDs (who have not yet been identified as such) would help bridge the treatment gap.
Using this information in practice, we believe that all colleges should offer cognitive dissonance-based group programming (the Body Project). Ideally these groups would be led by trained counselors, but training peer leaders was also proven to reduce ED risk factors, and in remote areas (distance-learning / places where they cannot recruit leaders) internet versions of the Body Project can be offered. Challenging what people believe about body image and eating habits changes their behaviors to remain consistent with new beliefs. By inviting them to publically (in a group setting) denounce the thin-ideal (or other body image fads) they set themselves up for creating positive changes.
*Key to abbreviations:
AIM = Affect Intensity Measure (Larsen, 1984)
BDI = Beck Depression Inventory (Beck, Steer, & Garbin, 1988)
BMI = Body Mass Index
BSQ = Body Shape Questionnaire (Cooper, Taylor, Cooper, & Fairburn, 1987).
CDI-SF = Child Depression Inventory – Short Form (Kovacs, 1992)
CES-D = Center for Epidemiologic Studies – Depression Scale (Radloff, 1977)
DERS = Difficulties in Emotion Regulation Scale (Gratz & Roemer, 2004)
DRES = Dutch Restrained Eating Scale (van Strien, Frijters, Van Stavern, Defares, & Deurenberg, 1986)
DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. (American Psychiatric Association, 1994)
DSM-5 = Diagnostic and Statistical Manual of Mental Disorders, Fifth edition. (American Psychiatric Association, 2013)
EAT-26 = Eating Attitudes Test (Garner, Olmstead, Bohr, & Garfinkel, 1982)
ED = Eating Disorder
EDDI = Eating Disorder Diagnostic Interview (Stice, Burton, & Shaw, 2004)
EDE = Eating Disorders Examination (Fairburn & cooper, 1993).
EDE-Q = Eating Disorder Examination-Questionnaire (Fairburn & Beglin, 1994).
EDDS = Eating Disorder Diagnostic Screen (Stice, Fisher, & Martinez, 2004)
EDI = Eating Disorder Inventory (Garner et al., 1983)
HWI = Healthy Weight Intervention (program) (Stice, Chase, Stormer, Appel, 2001; Stice, Shaw, Burton, & Wade 2006)
IBSS-R = Ideal Body Stereotype Scale-Revised (Stice, Ziemba, Margolis, & Flick, 1996).
PANAS-X = Positive Affect and Negative Affect Scale-Revised (Watson & Clark, 1992)
PSPS = Perceived Sociocultural Pressure Scale (Stice, Presnell & Spangler, 2002)
PSPS^ = Perceived Sociocultural Pressure Scale (Stice, Ziemba, Margolis, & Flick, 1996)
SATAQ-3 = (Sociocultural Attitudes Towards Appearance Questionnaire – 3 (Thompson et al., 2004)
SD-BPS = Satisfaction and Dissatisfaction with Body Parts Scale (Berscheid, Walster, & Bohrnstedt, 1973)
TOSCA-3 = Test of Self-Conscious Affect-3 Scale (Tangney, Dearing, Wagner, & Gramzow, 2000)
1. Stice E, Rohde P, Durant S, Shaw H. A Preliminary Trial of a Prototype Internet Dissonance-Based Eating Disorder Prevention Program for Young Women with Body Image Concerns. Journal of Consulting and Clinical Psychology. 2012;80(5):907-916. doi:10.1037/a0028016.
2. Black Becker, C, Bull, S, Smith, L, Ciao, A. (2008). Effects of Being a Peer-Leader in an Eating Disorder Prevention Program: Can We Further Reduce Eating Disorder Risk Factors? Eating Disorders.16(5), pp.444-459.
3. Stice E, Rohde P, Shaw H, Gau J. An Effectiveness Trial of a Selected Dissonance-Based Eating Disorder Prevention Program for Female High School Students: Long-Term Effects. Journal of Consulting and Clinical Psychology. 2011;79(4):500-508. doi:10.1037/a0024351.
4. Linville D, Cobb E, Lenee-Bluhm T, López-Zerón G, Gau JM, Stice E. Effectiveness of an Eating Disorder Preventative Intervention in Primary Care Medical Settings. Behaviour Research and Therapy. 2015;75:32-39. doi:10.1016/j.brat.2015.10.004.
5. Levitt DH. Participation in Athletic Activities and Eating Disordered Behavior. Eating Disorders. 2008;16(5):393-404. doi:10.1080/10640260802370556.
6. Müller S, Stice E. Moderators of the Intervention Effects for a Dissonance-Based Eating Disorder Prevention Program; Results from an Amalgam of Three Randomized Trials. Behavior Research Therapy. 2013;51(3):128-133. doi:10.1016/j.brat.2012.12.001.
7. Stice E, Durant, S, Rohde P, Shaw H. Effects of a Prototype Internet Dissonance-Based Eating Disorder Prevention Program at 1- and 2-year Follow-up. Health Psychology. 2012:77-86. doi:10.1093/med:psych/9780199859245.003.0007.
8. Gupta S, Rosenthal MZ, Mancini AD, Cheavens JS, Lynch TR. Emotion Regulation Skills Mediate the Effects of Shame on Eating Disorder Symptoms in Women. Eating Disorders. 2008;16(5):405-417. doi:10.1080/10640260802370572.
9. Stice E, Rohde P, Durant S, Shaw H, Wade E. Effectiveness of Peer-Led Dissonance-Based Eating Disorder Prevention Groups: Results from Two Randomized Pilot Trials. Behaviour Research and Therapy. 2013;51(4-5):197-206. doi:10.1016/j.brat.2013.01.004.
10. Becker CB, Mcdaniel L, Bull S, Powell M, Mcintyre K. Can We Reduce Eating Disorder Risk Factors in Female College Athletes? A Randomized Exploratory Investigation of Two Peer-Led Interventions. Body Image. 2012;9(1):31-42. doi:10.1016/j.bodyim.2011.09.005.
11. Stice E, Rohde P, Butryn ML, Shaw H, Marti CN. Effectiveness Trial of a Selective Dissonance-Based Eating Disorder Prevention Program with Female College Students: Effects at 2- and 3-year Follow-Up. Behaviour Research and Therapy. 2015;71:20-26. doi:10.1016/j.brat.2015.05.012.
12. Greif R, Becker CB, Hildebrandt T. Reducing Eating Disorder Risk Factors: A Pilot Effectiveness Trial of a Train-the-Trainer Approach to Dissemination and Implementation. International Journal of Eating Disorders. 2015;48(8):1122-1131. doi:10.1002/eat.v48.8.
13. Back To School Statistics. National Center for Education Statistics Website. https://nces.ed.gov/fastfacts/display.asp?id=372 Published 2017. Accessed November 11, 2017.