Eating Disorder Prevention Programs: A Meta-Analytic Review

1
Eating Disorder Prevention Programs A
Meta-Analytic Review

• Eric Stice and Heather Shaw
• Presented by Lisa Hoekzema
• Radford University

2
Purpose

• Better prevention programs need to be developed
  because
• Eating pathology is one of the most prevalent
  psychiatric problems for women and girls
• Less than a third of individuals with eating
  disorders receive treatment
• Treatment produces symptom remission in only 40
  to 60 of patients

3
Issues to Consider

• Will programs that are targeted at a high-risk
  group prove more successful than universal
  programs?
• Are intervention effects stronger for female only
  groups verses male or mixed gender groups?
• Do programs targeted at participants who are in
  the developmental period in which the pathology
  typically begins show better results?

4
Issues to Consider

• Are interactive programs more effective than
  didactic programs?
• Will stronger effects be seen in multiple session
  interventions relative to single session?
• Do programs that focus on established risk
  factors rather than non-established factors have
  a larger effect?

5
Dependent Measures

• Knowledge
• Body Mass
• Thin-ideal internalization
• Body dissatisfaction
• Dieting
• Negative affect
• Eating pathology

6
Literature Review

• Computer search (1980-2003)
• PsychINFO
• MEDLINE
• Dissertation Abstracts International
• Cumulative index to nursing and Allied Health
  Literature
• Key word Eating disorder, eating pathology,
  anorexia, anorexic, bulimia, bulimic, binge
  eating, prevention, preventive, intervention

7
Literature Review

• Table of contents from related journals
• International Journal of Eating Disorders,
  Journal of Consulting and Clinical Psychology
• Reference sections of all identified articles,
  chapters and books
• Researchers unpublished work
• Found Studies that met criteria
• 60 separate effect sizes

8
Inclusion/Exclusion Rules

• Included both trials that tested for intervention
  effects on eating pathology and those that
  focused on intervention effects on risk factors
  that have been established as predictors of
  eating pathology
• Must have random assignment to group and control
  group
• Must know if experimental group effects were
  significantly different from the control group
  effects
• Must include a minimum of 10 trials

9
Potential Moderators

• Target audience
• Method of presentation
• Age of participants
• Gender composition of group
• Number of sessions
• Program content
• Validity of scales used

10
Results Knowledge

• Average Effect Size at termination .30
• ?2 (15, N 16) 190.52
• Moderators none
• Average Effect Size at Follow-Up .29
• ?2 (12, N 13) 97.76
• Moderators none

11
Results Body Mass
Average Effect Size at termination .12 ?2 (10,
N 11) 25.95 Average Effect Size at Follow-Up
.05 ?2 (10, N 11) 18.26, ns
  Z _at_ termination Z _at_ Follow-UP
Group risk    
format 2.15  
age    
gender    
number of sessions    
content    
valid measures?    
12
Results Thin-Ideal Internalized

• Average Effect Size at termination .30
• ?2 (15, N 16) 190.52
• Average Effect Size at Follow-Up .15
• ?2 (16, N 17) 25.88, ns

  Z _at_ termination Z _at_ Follow-UP
Group risk 2.93  
format    
age 2.16  
gender 2.04  
number of sessions    
content -2.20  
valid measures? 2.50  
13
Results Body Dissatisfaction

• Average Effect Size at termination .13
• ?2 (45, N 46) 85.02
• Average Effect Size at Follow-Up .12
• ?2 (36, N 37) 76.11

  Z _at_ termination Z _at_ Follow-UP
Group risk 4.71 6.72
format 1.97  
age 2.03 3.81
gender   2.52
number of sessions 2.15  
content 3.52  
valid measures?   2.69
14
Results Dieting

• Average effect size at termination .11
• ?2 (32, N 33) 59.76
• Average effect size at follow-up .11
• ?2 (31, N 32) 66.58

  Z _at_ termination Z _at_ Follow-UP
Group risk 2.88 6.40
format 2.54 2.69
age   3.77
gender    
number of sessions 3.27  
content    
valid measures?    
15
Results Negative Affect

• Average effect size at termination .14
• ?2 (33, N 32) 77.41
• Average effect size at follow-up .09
• ?2 (22, N 23) 27.04, ns

  Z _at_ termination Z _at_ Follow-UP
Group risk 6.20  
format 2.28  
age 2.90  
gender    
number of sessions    
content -3.40  
valid measures?    
16
Results Eating Pathology

• Average effect size at termination .12
• ?2 (38, N 39) 83.11
• Average effect size at follow-up .12
• ?2 (27, N 28) 42.81

  Z _at_ termination Z _at_ Follow-UP
Group risk 5.34 4.80
format 3.93 1.99
age 2.55 2.93
gender    
number of sessions 2.48  
content -2.60 -2.04
valid measures?    
17
Conclusions/Recommendations

• There may be multiple approaches that prevent
  eating pathology
• Better outcomes for
• High-risk participants
• Interactive programs
• Participants who were 15 years old
• Female only programs
• Multiple sessions
• Studies that used validated measures
• If the best intervention was used, 60,157 cases
  of pathology could be prevented