A Cluster Randomised Controlled Trial of Three School-based Body Image Interventions Delivered by Teachers
A Cluster Randomised Controlled Trial of Three School-based Body Image Interventions Delivered by Teachers
Body image is one of the leading concerns for young people. Such concerns can have serious health consequences, including unhealthy weight control and exercise behaviours, depression, smoking, low self-esteem and misusing drugs and alcohol. Leading and emerging approaches for improving body image are effective among older adolescent females, and when delivered by expert providers (e.g., psychologists). However, there is a need for body image programmes that include males and so can be delivered in co-educational settings, and can be delivered by non-expert providers (e.g., teachers within schools) to facilitate cost-effective and ongoing use.
This project will evaluate three classroom-based programmes for improving young people's body image, delivered by class teachers to girls and boys aged 13-14 years in secondary schools. Two programmes are based on leading (cognitive dissonance) and emerging (mindfulness) approaches for improving body image, but which have not been previously evaluated with both boys and girls and when delivered by teachers. The third programme has previously demonstrated effectiveness in improving body image when delivered by teachers to boys and girls in the classroom setting, and therefore will provide a useful comparison to benchmark these programmes.
To assess the impact of each programme, the current study will compare the body image and well-being of students who take part in each programme to students who take part in an alternative programme, and those who do not take part in any programme (classes as usual), over a 6-month period. The investigators will also assess the extent to which gender and pre-existing levels of body image concerns impacts the benefit received from taking part in the lessons. To undertake this project, the investigators will recruit students and teachers from 24 schools in Bristol and surrounding areas in the South West of England. Students will complete questionnaire assessments of body image and well-being before and after the 5-week programme period, and again 6 months later to assess longer-term benefits. The investigators will also gain in-depth feedback from students and teachers via focus groups and interviews, in order to inform future improvement of the programmes for wider implementation.
Therefore, this trial will assess whether the two newly adapted programmes result in improved body image compared to those who receive their classes as usual, and compared to an existing teacher-led programme.
Background:
Body image is recognised as an important public health issue by governments, policy-makers, and health professionals. Poor body image is associated with engagement in unhealthy weight loss behaviours, depression, reduced academic performance, and has been established as an important risk factor in the development eating disorders. Within Western societies, poor body image is also common, with an estimated 25-61% of adolescent girls and boys being dissatisfied with their appearance. This underlines the importance of prevention and early intervention. As such, researchers, governments and health professionals are increasingly seeking effective evidence-based body image interventions that can be disseminated to the public, in a bid to reduce body image concerns and other adverse negative health related effects such as eating disorders. However, widespread dissemination of these programmes remains a challenge due to the difficulty of reaching individuals at risk, the traditional focus on females only, and the reliance on expert intervention facilitators.
Although it can be difficult reaching those at risk with individual targeted prevention programs, school-based interventions provide a useful opportunity to reach large and diverse groups of children and young people, crossing all levels of risk and gaining access to those who may otherwise not receive necessary intervention. Effective school-based interventions are designed to reduce the presence or influence of risk factors for body image concerns (e.g., internalisation of appearance ideals, appearance pressures from friends, family and the media, appearance-related teasing and bullying, and negative emotions such as depression and anxiety) in an effort to promote healthy body image. Previous school-based work has demonstrated some success in reducing body concerns at the conclusion of program implementation and over meaningful follow-up periods, with evidence showing media literacy and cognitive dissonance approaches to be useful in this context. However, there is still room for improvement in the magnitude of effects for universal programmes, and the majority are still targeted at girls and led by expert facilitators. Consequently, further evaluation of different approaches in this setting is required.
In addition, another factor which is still rather mixed within the literature, and has an impact on how universal programmes are delivered, is regarding the use of expert-led versus teacher-led interventions. Previous reviews have shown that programs are more effective when facilitated by dedicated expert interventionists. However, evidence from previous studies conducted by the investigators supports the notion that teachers can deliver a body image intervention effectively with minimal training. Hence, more research is needed to understand the effectiveness of teacher led interventions further, as successful task-shifting to non-expert providers has been identified as an important goal for disseminating mental health interventions at scale.
Therefore, the aim of this research is to evaluate the efficacy and acceptability of adapted interventions based on mindfulness and cognitive dissonance, respectively, for delivery to adolescent girls and boys (aged 13-14 years) by trained teachers. They will be compared against an existing universal intervention with support for teacher-led delivery and a class-as-usual control group.
Design:
This study will be a parallel 4-arm cluster randomised controlled trial. Schools will be randomly allocated to receive either the cognitive dissonance, mindfulness, or Dove Confident Me interventions, or classes-as-usual (control). The intervention duration is the same for all experimental conditions, and students will be assessed on body image and related outcomes at baseline, post-intervention, and 6-month follow-up (with 12-month planned pending further funding).
Participants, recruitment, and allocation:
Sample size for our repeated measures design was determined a priori. Assuming a small effect size of Cohen's d = .2 based on previous research of teacher-led programmes, a medium intra-individual correlation between repeated measures of .5, and setting power at .80, the sample size required to detect significance at the 5% level under individual randomisation was 294 students per arm. Applying an inflation factor to account for school level clustering based on a small intraclass correlation coefficient (ICC) of .01 (informed by previous studies), increased this requirement to 585 students per arm. Based on an average of 100 students from Grade 9 classes per school, the investigators aim to recruit 6 schools per arm. Thus the investigators aim to recruit 24 secondary schools in total (N = 2400 students) from Bristol and surrounding areas (UK).
Schools will be initially invited to take part via email and phone, and via involvement in a wider programme aimed at facilitating greater participation in higher education run by the Widening Participation team at the host institution. Allocation will be undertaken using restricted randomisation based on minimisation in order to achieve balance in each of the arms with regard to key school characteristics (geographical region, socioeconomic status, proportion of students learning English as a second language). This is the recommended approach for cluster randomised controlled trials (RCTs) that recruit and allocate sequentially, and will be implemented using freely available computer software (Minim).
Interventions:
Dove Confident Me
Dove Confident Me represents an existing school-based intervention that targets multiple risk factors for body dissatisfaction, addressing the nature and source of societal appearance ideals, media literacy, appearance-related social comparisons, appearance-related conversations and teasing, and encouraging 'body activism' to challenge appearance pressures. It consists of five 45-minute interactive lessons, designed to be delivered by teachers in school classrooms. The intervention was an adaptation of an existing evidence-based expert-led programme, Happy Being Me. Standardised materials include a detailed lesson plan for teachers, PowerPoint slides, video clips, and student activity sheets. Intervention materials are freely available online in five languages (selfesteem.dove.com).
Cognitive dissonance
Cognitive dissonance is conceptualised as the psychological discomfort stemming from conflicting cognitions and behaviours. Inducing such a state can be used to shift targeted attitudes and behaviours, as individuals are motivated to do so order regain equilibrium. In the context of body image and eating disorder prevention, dissonance-based interventions (DBIs) encourage a voluntary counter-attitudinal stance against socially prescribed appearance ideals in order to produce a conflict with, and therefore reduce, internalization of such ideals and their consequences. An existing school-based dissonance intervention which was previously tested among early adolescent girls with psychologist delivery has been adapted for delivery to mixed-gender adolescents by trained teachers. It has been designed to follow the same format and length as Dove Confident Me, and therefore consists of 5 x 45minute lessons, with detailed lesson plans for teachers, accompanying Powerpoint presentations and student activity sheets.
Mindfulness
Mindfulness refers to the awareness and acceptance of present-moment experience. As a psychological technique, it is growing in support for treating a range of mental health concerns. Mindfulness based interventions (MBI) are the next potential alternative to cognitive dissonance interventions (CDI) as they represent a potentially useful strategy for promoting positive body image. In this context, MBIs aim to both increase the capacity to refrain from automatic responses when confronted with appearance ideals and related sociocultural pressures, and reduce the intensity and impact of negative experiences if and when they do occur. An existing school-based mindfulness intervention which the investigators have previously tested among older adolescent girls with psychologist delivery has been adapted for delivery to adolescent girls and boys by trained teachers. It has been designed to follow the same format and length as Dove Confident Me, and therefore consists of 5 x 45minute lessons, with detailed lesson plans for teachers, accompanying Powerpoint presentations and student activity sheets.
Procedure Informed active consent will be obtained from school senior management (during recruitment), informed assent from parents (prior to baseline assessment), and informed consent from students (at baseline assessment). Prior to lesson delivery, school teachers will be trained in their allocated intervention by expert body image researchers (2 x 1.5hr sessions), and will be provided with all teaching materials necessary for delivering the interventions. Students will complete questionnaires under standardised conditions supervised by their teachers and trained research staff. Approximately one week following baseline questionnaires, intervention schools will deliver their allocated intervention, facilitated by schoolteachers to whole classes (one lesson per week over five weeks), with a ratio of one teacher to 20-30 students. All lessons will be audio-recorded and a selection of lessons will be observed live by trained research staff, for the purposes of assessing fidelity to the intervention manuals. Teachers and selected students will take part in focus group interviews following completion of the lessons, in order to obtain qualitative feedback. Students in control schools will take part part in their usual scheduled lessons. All students will then complete follow-up questionnaires at post-intervention (within one week of final lesson), and 6-month follow-up. Blinding of students, teachers, and researchers is not possible due to the nature of the interventions. However, risk of bias from teachers and researchers is minimised due to anonymous self-report assessments. Risk of contagion is also minimised by randomising at the school level. Participating schools will receive a £200 honorarium upon project completion. This project has received ethical approval from the University of the West of England's Faculty Research Ethics Committee (HAS.16.09.016).
Statistical analysis plan:
Multiple imputation will be used to handle missing data. Analysis of intervention effects will be conducted using an intent-to-treat approach, with a sensitivity analysis conducted depending on the extent of missing data. Multilevel mixed models will be used to compare groups at post-intervention and 6-month follow-up on primary and secondary outcomes, accounting for the clustered nature of the design (repeated measures within students, within classes, within schools). Baseline levels of outcomes, as well as demographic and school characteristics will be used as covariates in order to increase precision of effects. Follow-up analysis will investigate moderation with respect to gender and baseline body image concerns, and will be conducted by including interactional terms in the statistical models. Subsequent mediational analyses following recommendations by Kraemer will investigate hypothesised mechanisms of change in each of the trial arms. An analysis of cost-effectiveness will also be undertaken, comparing relative effectiveness across interventions.
Inclusion Criteria (school):
Exclusion Criteria (school):