Elsevier

Preventive Medicine

Volume 52, Issue 2, 1 February 2011, Pages 152-158
Preventive Medicine

Real-world effectiveness of a German school-based intervention for primary prevention of anorexia nervosa in preadolescent girls

https://doi.org/10.1016/j.ypmed.2010.11.022Get rights and content

Abstract

Objective

Anorexia nervosa (‘AN’) is notoriously difficult to treat, has high mortality rates, and has a prevalence peak in 15-year-old girls. We developed a German school-based intervention program (‘PriMa’) for the primary prevention of AN in preadolescent girls and assessed the effects in a sample of Thuringian girls.

Method

Intervention involved nine guided lessons with special posters and group discussions. A parallel controlled trial with pre-post measurements and a three-month follow-up was conducted in 92 Thuringian schools (n = 1553 girls) in 2007 and 2008. Primary outcomes were conspicuous eating behavior, body self esteem, and AN-related knowledge.

Results

After adjusting for the girls' ages and the type of school, we observed significant improvements in the areas of knowledge (d = .24) and body self esteem (d = .29), but not for eating behavior.

Conclusion

The PriMa intervention provides an efficient and practical model to increase AN-related protection factors.

Introduction

Anorexia nervosa (‘AN’) is one of the most serious psychiatric disorders in girls and young women. The Diagnostic and Statistical Manual of Mental Disorders (DSM IV-TR) includes four criteria for diagnosing AN (see Table 1). Because it is difficult to treat (Steinhausen, 2002) and because of its high mortality rate (> 15%; Zipfel et al., 2000), the disease has gained much attention from the public and from the scientific community. AN's prevalence is estimated at 0.5 to 3.7%, with a peak of incidence at age 15 (Hoek and van Hoeken, 2003, Bulik et al., 2006). Several of AN's symptoms are considerably more common (Grilo, 2006), which are known as sub-clinical expressions of AN. In view of this, groups around the world have spent decades developing prevention programs to prevent the onset and consequences of AN and other eating disorders (for review, see Levine and Smolak, 2006).

According to the standards of the Society of Prevention Research (‘SPR’), a successful prevention approach includes the demonstration of efficacy (level 1) as well as the probation in the field under real-world conditions (level 2 ‘effectiveness’). ‘Real-world conditions’ mean regular conditions in everyday life without the motivational character of a new program introduced for the first time. Furthermore, an appropriate dissemination is necessary to secure sustainability (level 3 ‘broad dissemination’; Flay et al., 2005).

Most of the existing prevention programs do not meet all of the standards, including level 2 and 3. For this reason, in 2003, our institution began cooperating with the Thuringian Ministry of Education (‘TME’) to develop the program called ‘Primary prevention of anorexia nervosa in preadolescent girls’ (‘PriMa’). Our program focuses exclusively on AN, because it occurs earlier (peak of incidence: 15 years) than bulimia nervosa and binge-eating disorder (peak of incidence: 18 years; Grilo, 2006). In 2004 and 2005, we conducted a pilot study to evaluate the process and impact of PriMa (level-1 evaluation; n = 1006 girls; mean age = 12 years; Berger, 2008). This study showed significant improvements in pre–post measurements and pre–follow-up measurement on body self esteem, eating behavior, and knowledge.

Furthermore, we paired with the Heidelberg Prevention Center (www.h-p-z.de; level-3-evaluation; see Berger et al., 2008) to create appropriate distribution structures. Two well-known meta-analyses (Stice and Shaw, 2004, Stice et al., 2007) have summarized the effects of over 80 international programs for the prevention of eating disorders. The authors clearly identify factors that successful programs exhibit:

  • interactive (vs. didactic),

  • selective, for risk groups (vs. universal),

  • multisession (vs. single-session),

  • gender specific, offered only to females (vs. co-educative),

  • offered to participants over age 15 (vs. younger participants), and

  • delivered by professionals (vs. endogenous providers like teachers).

Because of the early peak of incidence of AN, we decided to focus PriMa on 6th-grade girls with a mean age of 12 years. According to most of the programs, PriMa is geared to the WHO-recommended ‘Setting Approach’ (Stewart-Brown, 2006), realizing the program in schools. We focused primarily on girls because adolescent girls and young women are 10-times more likely to exhibit incidents of AN (Grilo, 2006). We adopted a universal prevention program to avoid labeling the participating girls who were already suffering from (sub-clinical) eating disorders or showing specific risk factors. Other universal prevention programs were very effective (Becker et al., 2005, Raich et al., 2009) and primary prevention by definition includes this universal character. Furthermore, our cooperation partners insisted using teachers to implement the project rather than independent professionals on the theory that teachers would improve the program's sustainability.

Existing prevention programs concerning eating disorders focused on decreasing risk factors (e.g. weight concerns, negative body image, dieting, and low social support; Jacobi et al., 2004) and strengthening protective factors (e.g. body self esteem, and knowledge). The present study aims to follow this primary preventive approach (Caplan, 1964) to reduce the occurrence of AN (see Table 2). Furthermore we intend to replicate the results of the pilot study concerning the effects of the PriMa program under real-world conditions (level 2; Berger et al., 2008). In this context, the focused research question is whether the primary preventive intervention program, PriMa, is able to significantly improve the primary outcomes eating behavior, body self-esteem, and AN-related knowledge in comparison to a control group?

Section snippets

Study design and subjects

We designed a controlled trial with (post-hoc) parallel group assignment using pre-post-design, supplemented by a follow-up-measurement after about three months. Before starting the program, we first obtained the consents of the parents, the supervisor of the TME, and the ethics committee of the Jena University Hospital. Our close cooperation with TME required several compromises because TME was focused on the practical effects of the program rather than the underlying research project. Because

Patient characteristics

The demographic characteristics of the girls (IG vs. CG) can be seen in Table 3. IG and CG did not differ significantly, except for ‘age’ (t(885) = 2.66, p = .01). However, this age difference (12.01 years vs. 11.95 years) has a low substantive importance with an effect of d = 0.1 (mean difference about 22 days). Potential distortions through hierarchical effects (individuals in grades in schools) could be ruled out, because the ICC determined from the data is 0.02 (Raudenbush and Bryk, 2002).

793 girls

Discussion

The pre–post control group study described above aimed to verify the effectiveness of PriMa against AN for girls attending the 6th grade that are an average of 12 years old. According to Table 4, we found that values for all variables in the IG developed in accordance with the hypotheses. Compared to the CG, a significant intervention effect could be found only for the knowledge at post as well as follow-up measurement. The body self esteem improved significantly only in the short term

Conclusion

Taking into account the positive results of the pilot study (Level-1, including process evaluation) and the successful implementation of the program in cooperation with the TME and the Heidelberg Prevention Center (www.h-p-z.de), PriMa is a girl-specific, school-based primary preventative method, and is therefore a low-threshold and socially fair intervention. With relatively little effort (9-hour project teaching) and low costs, the program is able to increase AN-specific knowledge and to

Conflict of interest statement

None.

Acknowledgments

We thank the financial support by the Federal Ministry for Education and Research (BMBF: Project no. 01EL0602, term 2006–2009).

Furthermore we thank our practice partner Thuringian Ministry of Education, especially Jutta Beinersdorf and Margit Luedecke for the didactic counseling and competent application of teacher trainings and all the girls for patiently filling out the questionnaires.

References (36)

  • J.L.R. Martin et al.

    Meta-analysis of drop-out rates in randomised clinical trials, comparing typical and atypical antipsychotics in the treatment of schizophrenia

    Eur. Psychiat.

    (2006)
  • S. Zipfel et al.

    Long-term prognosis in anorexia nervosa: lessons from a 21-year follow-up study

    Lancet

    (2000)
  • C.B. Becker et al.

    Reducing eating disorder risk factors in sorority members: a randomized trial

    Behav. Ther.

    (2005)
  • U. Berger

    Essstörungen wirkungsvoll vorbeugen (Effective Prevention of Eating Disorders)

    (2008)
  • U. Berger et al.

    Primary prevention of eating disorders: characteristics of effective programs and how to bring them to broader dissemination

    Eur. Eat. Disord. Rev.

    (2008)
  • B. Buddeberg-Fischer et al.

    Prevention of disturbed eating behavior: an intervention program in Swiss high school classes

    Eat. Disord.

    (2001)
  • C.M. Bulik et al.

    Prevalence, heritability and prospective risk factors for anorexia nervosa

    Arch. Gen. Psychiat.

    (2006)
  • G. Caplan

    Principles of Preventive Psychiatry

    (1964)
  • J.C. Carter et al.

    Primary prevention of eating disorders: might it do more harm than good?

    Int. J. Eat. Disord.

    (1997)
  • J. Cohen

    Statistical Power Analysis for the Behavioral Sciences

    (1988)
  • S.M. Eldridge et al.

    Sample size for cluster randomized trials: effect of coefficient of variation of cluster size and analysis method

    Int. J. Epidemiol.

    (2006)
  • E. Erdfelder et al.

    GPOWER: a general power analysis program

    Behav. Res. Meth. Ins. C.

    (1996)
  • C.G. Fairburn et al.

    The Eating Disorder Examination

  • A.E. Field et al.

    The validity of self-reported weight change among adolescents and young adults

    Obesity

    (2007)
  • B.R. Flay et al.

    Standards of evidence: criteria for efficacy, effectiveness and dissemination

    Prev. Sci.

    (2005)
  • D.M. Garner et al.

    The Eating Attitudes Test: an index of the symptoms of anorexia nervosa

    Psychol. Med.

    (1979)
  • M. Gerlinghoff et al.

    Magersucht und Bulimie

    Verstehen und Bewältigen (Anorexia nervosa and bulimia nervosa: Understanding and Coping)

    (1993)
  • M.G. Grilo

    Eating and Weight Disorders

    (2006)
  • Cited by (0)

    1

    Fax: +49 3641 936546.

    View full text