Body image concerns are pervasive and can impact multiple aspects of a person's life. Individuals with negative body image tend to have negative thoughts and feelings about their bodies, often resulting in unhealthy behaviours. Negative body image is also associated with mental and physical health conditions. Thus, prevention efforts are crucial to alleviating body image concerns in young adults as they are considered high-risk populations.

This study is a randomised controlled trial that aims to evaluate the effectiveness of a self-guided body image program on a mobile-based application for young adults.

Hypothesis 1a: The intervention group will improve significantly on measures of body image at post-intervention and follow-up, as compared to the control group.

Hypothesis 1b: The intervention group will reduce significantly on measures of appearance-ideal internalization, media and peer pressures at post-intervention and follow-up, as compared to the control group.

Hypothesis 2: The intervention group will improve significantly on the measure of self-compassion at post intervention and follow-up, as compared to the control group.

Procedure:

Participants will begin by reading the Participant Information Sheet (PIS) online. Once they have indicated consent on the online Consent Form, participants will proceed to complete a set of questionnaire on an online survey platform (Qualtrics) to obtain baseline ratings. The online questionnaire consists of four measures on body image, body image risk factors and self-compassion. Participants will also report demographic information such as gender and age.

Next, participants will download the phone application and be guided to navigate the application. The phone application for the programs in this study is provided by Intellect Pte Ltd, a Singaporean technology start-up company that collaborates with the supervisor of this project (OS). All participants will be randomized to one of two conditions- intervention or active wait-list control-using simple randomisation procedures. Participants in the intervention condition will participate in 9 days of body image program, while participants in the active wait-list control condition will participate in 9 days of cooperation learning program. In both programs, participants will engage in content learning and short activities daily on the phone application. The anticipated time participants will spend on each programs is comparable, of about less than five minutes per day.

Participants will be asked to complete the same set of online questionnaires immediately after completing the program (post-intervention measure), and four weeks after completing the program (follow-up measure). The feedback questionnaire will be administered only post-intervention. Participants will receive either course credit or a small monetary reimbursement after completing the program and questionnaire at post-intervention and at the four weeks follow-up.

At the end of the six weeks data collection period, all participants will be debriefed about the purpose of the study. Participants in the active wait-list control condition will be given access to the body image program.

Outcome Measures:

Body image measures:

Body Image Ideals Questionnaire (BIQ) is a 22-item scale that assesses body image satisfaction-dissatisfaction by measuring the degree of congruence or discrepancy of one's perceived and idealized physical attributes. On a scale from 0 (Exactly as I am) to 3 (Very unlike me), participants rate the degree they resemble their physical ideal on 11 physical attributes. Next, participants rate the importance that they attain their ideal on each physical attribute. The cross-products of the discrepancy and importance ratings for each physical attribute will be obtained and a composite BIQ score will be computed. Higher scores indicate a greater disparity between one's perceived and ideal physical attributes, suggesting higher levels of body dissatisfaction. BIQ has good internal consistency, with Cronbach's α of 0.81 for males and 0.76 for females.

Body Appreciation Scale-2 (BAS-2) is a 10-item scale that assesses individuals' positive attitudes toward their bodies. Items are scored on a scale from 1 (Never) to 5 (Always). Scores on all items are averaged with higher scores indicating greater body appreciation. BAS-2 has excellent internal consistency, with Cronbach's α of 0.96 for males and 0.97 for females.

Risk factor measures:

Sociocultural Attitudes Toward Appearance Questionnaire-4 Revised (SATAQ-4R) measures internalization ideals and appearance-related sociocultural pressures. The seven subscales consist of 31 items for females and 28 items for males on a scale from 1 (definitely disagree) to 5 (Definitely agree). Higher scores on each subscale indicate higher levels of internalisation and sociocultural pressures. The internal consistency of the subscales are good, with Cronbach's α of 0.82 or higher in a sample of university females and Cronbach's α of 0.75 or higher in a sample of university males.

Self-compassion measure:

Self-Compassion Scale-Short Form (SCS-SF) is a 12-item scale that measures self-compassion on six subscales. Each item is scored from 1 (Almost never) to 5 (Almost always). A total self-compassion score is a grand mean of all six subscale means, with higher scores indicating higher levels of self-compassion. The internal consistency of the scale is excellent, with Cronbach's α of 0.86.

Feedback questionnaire:

App Engagement Scale is a 7-item scale that measures participants' engagement on the phone application form 1 (Definitely disagree) to 5 (Definitely agree). A total score is derived by adding scores from each item. The internal reliability of the scale is good, with Cronbach's α of 0.84.

Power analysis and sample size:

Most studies that examined mobile and internet-based body image and eating disorder programs reported intervention effects of small to moderate effect sizes. Using G*power, a small effect size of Cohen's f = 0.1, and alpha = 0.05, an estimated sample size of 214 participants are required to achieve a power of at least 0.90. Taking into consideration a potential attrition rate of 30% in app-based interventions, 280 participants are required.

Statistical analyses:

Statistical analyses will be done using SPSS Version 26.0. All analysis will be conducted using two-tailed test with a level of significance at p < .05.

Data will first be visually inspected using scatterplots, histograms and boxplots to examine the distribution of data and identify significant outliers. Data further than three standard deviations from the mean will be identified as significant outliers. Further analyses will be conducted with and without the outliers to determine if outliers influence the results significantly. If outliers influence results significantly, they will be managed using the Winsorizing method, where outliers will be replaced with the next closest non-outlying data point.

A one-way Analysis of Variance (ANOVA) will examine baseline differences between the two conditions on all demographic and dependent variables. If no significant differences are observed at baseline between the two conditions, two-way 2 (condition) X 3 (time) mixed factorial repeated measures ANOVA will be conducted to determine if there are significant changes over time in the intervention and active wait-list control conditions. Any significant interaction terms (condition x time) will be examined using simple slope analysis. Tests of statistical assumptions such as Mauchly's Test of Sphericity and Levene's test for homogeneity of variances will be carried out before the main analyses.

If significant differences are observed at baseline between the two conditions, Analysis of Covariance (ANCOVA) will be conducted, controlling for baseline values. Tests of statistical assumption for ANCOVA such as independence between covariate and independent variable (condition), and homogeneity of regression slopes will be carried out.

For both analyses, if participants in the intervention condition improved significantly on the outcome measures at post-intervention and follow-up compared to participants in the active wait-list control condition, it provides evidence that the intervention is effective in improving body image and self-compassion in university students.

Intent-to-treat analysis will be carried out to address the loss of participants' data due to attrition, by carrying forward the participants' last reported score. Analyses using one-way ANOVA on demographic and dependent variables will be done to determine if there are any significant differences between participants who dropped out and participants who did not. Any significant differences detected can inform of attrition-related biases in the study.

Inclusion Criteria:

• Age 18-30
• good command of the English language

Exclusion Criteria:

- none

Research data will be stored in an encrypted NUS nbox folder with viewing-only permissions set and restricted to the PI and co-investigator. Primary data will be kept secured in electronic form by the PI and co-investigator in a format that precludes subsequent alteration. Research data files (e.g Excel sheets) will only be accessible by PI and Co-PI. Survey data will be collected on Qualtrics secured with an account password and 2FA verification. Upon completion of the data collection, personal identifiable information(e.g. matric number from RP pool students) will be removed from the Qualtrics file, and the MS Excelsheet containing the data will be encrypted and stored in an encrypted NUS nbox folder, accessible only by the PI and Co-investigator. App data will be linked to a participant code given to participants. However, no personal identifiable data will be collected through the wellbeing app programs. Personal data will be only be shared with PI and Co-I as mentioned above.