Women in general and young girls in particular are constantly exposed to unhealthy body and appearance ideals through media that contribute to body dissatisfaction and unhealthy behaviors such as rigid dieting, which in interaction with genes and other factors increase the risk of developing eating disorders such as anorexia nervosa and bulimia nervosa. The investigators aim is to investigate the extent to which an interactive prevention program, delivered through Internet, called the Swedish Body Project (sBody Project) can decrease the emergence of eating disorders among young females. The sBody Project is based on a "Dissonance-Based Intervention: (DBI)" that has shown very promising results. The adaptations and changes in the format of delivery accomplished in this study might help to disseminate the program on a broad basis, and consequently affect the health of young females on a much larger scale the ever before.

OVERALL AIM AND RESEARCH QUESTIONS The overall aim of this project is to adapt the DBI to be implementable on a large scale by means of interactive groups working online, and new methods for easily and effectively training program facilitators), with recruitment using the model of social media (e.g., through various websites, collaborative projects, and blogs that are popular among the youth), and to evaluate the effect of the adapted version of the DBI. The program will remain interactive, and its effect can most probably be maintained over time to a higher degree than when implemented in its original format, as booster sessions can easily be included with no major problems in terms of logistics, meetings, allocation of participants, etc. The new format will also provide a framework for recruitment which reduces the risk of stigmatization.

Research questions

1. How effective is the Internet-based DBI, immediately after the intervention and 6, 12, 18, and 24 months later, to reduce the onset of eating disorders?
2. What channels in the Internet / social media leads to the best recruitment of the target group?
3. Can further use and implementation of Internet-based DBI be maintained through a collaborative project (e.g., Wikipreventia) where the principles and methods of implementation (from the training of facilitators to recruitment, implementation and evaluation) are described, facilitated, and the experiences are accumulated?
4. Is the adapted DBI cost-effective?
5. Secondary questions concern the new format's potential for engaging the participants in booster sessions, degree of stigma in recruitment, mediators and moderators of outcome, and subgroup analyses.

STUDY DESIGN, MATERIAL AND METHODS Study Design The DBI will be adapted to be widely implemented and delivered via Internet and through a mobile "app", while maintaining its interactivity and underlying principles and theories. An initial evaluation of this model, although in a simpler format than what is proposed here, has shown promising results.

In a randomized trial, a total of at least 400 girls aged 15-18 with self-reported high levels of body dissatisfaction that is an established risk factor for the development of eating disorders will be randomized into three groups:

1. Internet-based DBI, which consists of four interactive occasions, some homework assignments, and monitoring,
2. Expressive Writing (placebo/attention control), and
3. A wait-list control condition.

Those in the wait-list condition will randomized into one of the active interventions as soon as they have done a 6-month follow-up assessment. Direct and long-term effects of the interventions (6, 12, 18, and 24 months post-intervention) will be investigated by systematic measurements.

Study population and sample size Following the ethical guidelines from the Swedish Research Council, informed consent will be obtained, and an initial assessment (telephone interview) will be conducted. Those who meet the diagnostic criteria for any eating disorder, body dysmorphic disorder, or other serious conditions that require psychiatric care will be excluded. A booster session will be available every six months.

With an effect size of d=.32 between the DBI and the active control condition regarding the main outcome variable (eating disorder pathology), with three conditions in total, and at least 3 measurements (pre, post and 6-months follow-up) the investigators need at least 260 participants with p < .05, an expected drop-out of 15%, and reliability of measures set between .75 and .95 to attain a power of at least .95. Given the recruitment channels, the investigators expect a much larger number than 260 to participate in the study, and to ensure adequate power for analyses using random effects and to take further drop-out over time into consideration the investigators aim for at least 400 participants.

Procedure Participants will be recruited through notes in different social media and the website of the project that will be linked to different fora which are known to engage adolescent girls. After obtaining informed consent, participants will be asked to download the "app" used for the study. Participants will then be asked to respond to the set of questionnaires (pre-intervention assessment) using a secure plattform, and they will be interviewed. They will then be asked to await feedback, and if included allocation to a group. Participants with significantly high scores on measures of depressive symptomatology or potential diagnosis of dysmorphophobia, eating disorders will be advised to seek professional help, and receive guidance on how to do it. The rest of the participants will be randomized to the study arms. They will then receive further information about their participations and use of the app.

The first session will be initiated by the facilitators who introduce basic issues about body image and the thin ideal.

The day before, and an hour before each meeting, the participants would receive a reminder.

The sessions will be each one hour long. The facilitators will guide the participants through the content of each session following a written manual. Participants who miss a session will be asked to listen to the recorded session ahead of the next session, and briefly discuss it with a facilitator.

Those randomized to Expressive Writing condition will be asked to reflect upon issues related to body image and self-concept according to a manual.

Participants will be asked to respond to the post-intervention assessment battery after the end of the fourth session. They will also be invited to participate in the booster sessions provided after the end of the intervention, and to respond to the follow-up assessments.

Inclusion Criteria:

• Must have self-reported body dissatisfaction that is an established risk factor for the development of eating disorders.

Exclusion Criteria:

• Those who meet the diagnostic criteria for any eating disorder diagnosis, body dysmorphic disorder, or other serious conditions (e.g., bipolar disorders, schizophrenia) that require psychiatric care will be excluded.