Participants and procedure

Study data were collected and managed using REDCap (Research Electronic Data Capture) tools hosted at the University of Wisconsin [25, 26]. In June 2021, recruitment emails with a link to an online screening survey were sent to all students enrolled in a medicine (MD) program at the University of Wisconsin. In June 2022, eligibility criteria were expanded to include students enrolled in one of four health professional programs at the university – MD, bachelor of science in nursing (BSN), physician associate (PA), or physical therapy (DPT) – in order to better capture individuals who will work in a variety of patient care settings. Study participants were considered eligible if they (1) were enrolled at time of recruitment in an MD, BSN, PA, or DPT program, (2) were between the ages of 20 and 45 years old, (3) had access to the internet and an electronic device with a webcam and a microphone, and (4) were available during session dates and times. The study and all changes of protocol were approved by the Institutional Review Board (IRB; ID: 2021 − 0564).

Eligible participants received the consent form online via REDCap. Those who provided informed consent were randomized to either an intervention condition consisting of two small-group, peer-led virtual BAM-Health sessions, or a control condition consisting of an informational brochure about weight stigma from the Obesity Action Coalition and the Yale Rudd Center for Food Policy and Obesity [27]. Participants assigned to BAM-Health were further divided into small groups of six to nine participants each. Once randomized, participants received emails with information about their assigned condition (i.e., how to access the BAM-Health sessions) and links to study measures at three different timepoints: at pre-intervention (baseline), immediately following intervention completion (post-intervention), and at four weeks following intervention completion (follow-up). Those assigned to the control condition were provided informational brochures via email on the date of the first BAM-Health session. All study participants received automated email reminders to complete the post-intervention and follow-up surveys within one week of receipt. Participants were compensated with $40 gift cards for completion of all assessments and BAM-Health sessions if assigned to the BAM-Health condition.

Intervention

BAM-Health is a group-based, peer-led intervention intended to target weight stigma on intrapersonal, interpersonal, and institutional levels. BAM-Health was adapted from the original Body Advocacy Movement (BAM) curriculum, which was designed to reduce eating disorder (ED) risk by addressing fear of weight gain and weight stigma among young adults [28]. BAM utilizes psychoeducation, cognitive dissonance, and exposure-based strategies to target participants’ attitudes and behaviors related to their own bodies. While BAM focuses primarily on challenging weight stigma to reduce ED risk among young adults, the BAM-Health curriculum includes additional emphasis on aspects of health and healthcare, encouraging health professional students to consider how weight stigma manifests not only in their personal lives, but also in patient care practices and in healthcare institutions. Adaptations to the BAM manual for use among medical trainees were led by medical student co-authors (SF, EB, TM, AK) and overseen by a clinical psychologist (KS). In two, two-hour sessions held one week apart, with three between-session activities, BAM-Health aims to critically examine weight stigma in health professional students’ personal and professional lives across all levels of the socioecological model (see Table 1).

Table 1 BAM-Health sessions outline, sample activities, sample discussion questions (italicized), and level(s) of the socioecological model targeted within each section

In 2022, BAM-Health peer facilitators held a compensated, voluntary focus group with five members of the medical school’s Student Assembly for Minority Concerns recruited by email to assess the BAM-Health curriculum’s cultural relevance, applicability, and utility for students of diverse backgrounds. Focus group participants did not formally participate in BAM-Health, but reviewed the intervention manual and provided feedback. Thereafter, minor changes to language were made to the intervention script (i.e., addition of a Diversity, Equity, and Inclusion acknowledgement in the intervention introduction, alteration of names and characters in role-play activities to reflect individuals of varied backgrounds). The overall composition of the BAM-Health intervention, in terms of discussion content and activities, remained the same between 2021 and 2022.

Training

Peer facilitators were MD students who underwent a 16-hour training program led by a clinical psychologist specializing in EDs, trained in Body Project and BAM facilitation, and who oversaw development of the BAM-Health manual (KS). During training, facilitators divided into two groups of two to three and practiced delivering and receiving the intervention and received constructive feedback, a training method supported in similar peer-facilitated programs [28]. In 2021, 4 peer facilitators were trained. In 2022, 5 peer facilitators were trained. Table 2 describes the components of the two-day training program in further detail.

Table 2 Sample BAM-Health facilitator training scheduleMeasuresDemographics

At baseline, participants provided demographic information including age, gender identity, race and ethnicity, sexual identity, and health professional program affiliation.

Measures targeting internalized weight and appearance concernsGoldfarb fear of fat scale (GFFS)

Participants completed the GFFS [29], a 10-item questionnaire that assesses respondents’ internalized fear of fatness, at all three assessment timepoints. Participants indicated the extent to which they believed statements were true on a four-point Likert scale from (1) “Very untrue” to (4) “Very true,” and responses were summed to create a total score. Sample items include “My biggest fear is of becoming fat” and “There is nothing that I can do to make the thought of gaining weight less painful and frightening.” The GFFS has demonstrated strong internal consistency (ɑ = 0.85), 1-week test-retest reliability (r = .88), and discriminant validity [29]. Internal consistency in the present sample was good (ɑ = 0.86).

Ideal-body stereotype scale-revised (IBSS-R)

Thin-ideal internalization was assessed with the IBSS-R [30], a 10-item, self-report scale which asks participants to rate items such as “Slender women are more attractive” and “Women with long legs are more attractive” on a five-point Likert scale from (1) “Strongly disagree” to (5) “Strongly agree.” Higher mean ratings across the 10 items indicate a greater degree of thin-ideal internalization. The IBSS-R has strong internal consistency among undergraduate women (ɑ = 0.91) [31], as well as test-retest reliability and convergent and predictive validity [32]. Participants completed the IBSS-R at all three assessment timepoints. Internal consistency in the present sample was good (ɑ = 0.85).

Measures targeting interpersonal weight stigmaUniversal measure of bias – fat version (UMB-FAT)

The UMB-FAT [33], a 20-item, four-factor scale, assesses respondents’ opinions of individuals in larger bodies. Participants indicate their agreement with each item on a seven-point Likert scale (for example, “I would be comfortable having a fat person in my group of friends”) from (1) “Strongly agree” to (7) “Strongly disagree.” Higher mean scores indicate greater bias. In a university sample, the UMB-FAT displayed strong internal consistency (ɑ = 0.87), inter-item correlation (0.25), and less susceptibility to social desirability response bias than comparable assessments [33]. Participants completed the UMB-FAT at all three assessment timepoints. Internal consistency in the present sample was good (ɑ = 0.91).

Eating pathology symptoms inventory – negative attitudes towards obesity subscale (EPSI-NATO)

The EPSI [34] is a comprehensive, 45-item eating pathology scale with eight validated subscales, including a subscale measuring “Negative Attitudes Towards Obesity” (NATO), which was analyzed in this study as a measure of interpersonal weight stigma. The NATO subscale includes five items assessing thoughts towards individuals in larger bodies. Participants are asked to consider each item in the context of the previous four weeks and indicate its frequency on a five-point Likert scale from (0) “Never” to (4) “Very often.” Responses are summed to create a score for the subscale. Sample EPSI-NATO items include, “I thought that obese people lack self-control” and “I felt that overweight people are lazy.” The EPSI has good convergent and discriminant validity, internal consistency, and two to four week test-retest reliability in clinical and university samples [34,35,36]. Participants completed the EPSI at all three assessment timepoints; however, only the EPSI-NATO is discussed here. Internal consistency for the EPSI-NATO subscale in the present sample was good (ɑ = 0.88).

Acceptability and feasibility

Acceptability and feasibility were assessed only in 2022. Given the pilot nature of this study, in 2021, research activities focused primarily on developing the BAM-Health manual and addressing initial challenges of implementation. In 2022, an added focus was assessing acceptability of the intervention from participants’ perspective and feasibility of implementation.

Acceptability

In 2022, feedback surveys were provided at post-intervention that evaluated overall enjoyment of the intervention, comfort in participating in group discussions, relevance of the discussions and information presented, readiness to apply BAM-Health concepts in daily life, perceptions of body image improvement, and quality of peer facilitators’ instruction. Participants rated their responses on a Likert scale from (1) “Strongly disagree” to (5) “Strongly agree.” Internal consistency was good (ɑ = 0.87).

Participants were also asked to provide written responses to several open-ended questions, including: (1) What did you like most about BAM Health? Do any particular activities or moments stand out for you? (2) What did you like least about BAM Health? Do any particular activities or moments stand out for you? (3) How did you feel about the Zoom environment? Feel free to share both positives and negatives. (4) Was there anything missing from BAM Health that you wish had been included? and (5) Do you have any other comments you would like to share?

Feasibility

Feasibility was assessed in 2022 through evaluation of BAM-Health facilitator competence and session adherence to the intervention manual and associated script. Sessions were recorded in 2022, and two independent raters (AL and MG) reviewed all session videos, assessing adherence and facilitator competence using scales adapted from prior trials of the Body Project, a similar peer-facilitated intervention [37], with content checks changed to reflect BAM-Health content. Session adherence (Appendix A) was rated on a Likert scale from 1 (“No adherence. This section was skipped entirely”) to 10 (“Perfect! Absolutely all material in the section was presented exactly as written”). Facilitator competence (Appendix B) was assessed based on pace, clarity, organization, ability to keep the group on task, ability to provide equal speaking time for all members of the group, listening skills, respectfulness, warmth, and enthusiasm, with each item graded on a ten-point scale from 1 (Poor) to 10 (Superior). Completion of between-session activities (Table 1) by participants was also tracked to assess feasibility. Participants anonymously uploaded their activities via an online form and submissions were rated based on general completion.

Analysis plan

Aim 1 was evaluated using baseline to post-intervention and baseline to follow-up effect sizes on internalized weight and appearance concerns and interpersonal weight stigma measures. Effect sizes were chosen rather than null hypothesis significance testing due to the limited power and pilot nature of the study. When comparing effect sizes, Cohen’s d absolute effect sizes greater than 0.20 were considered small, greater than 0.50 were considered moderate, and greater than 0.80 were considered large [38].

Acceptability of BAM-Health (Aim 2) was assessed via attrition rates across the two intervention sessions, with the goal of attaining attrition rates of less than or equal to 20%. Average ratings from post-intervention feedback surveys were calculated for each item, with the goal of attaining an average rating of greater than or equal to four out of five (80%) on each item. A thematic analysis of free responses to open ended questions at post-intervention was conducted to elucidate main themes. To assess feasibility (Aim 2), average ratings for facilitator competence and session adherence were determined, with the goal of achieving average ratings of greater than or equal to eight out of 10 (80%). To calculate interrater reliability, percent agreement of scores were compared with a goal of agreement between the two reviewers of greater than or equal to 80%. Feasibility was also assessed via completion rates of between-session activities and comparison of attrition rates between BAM-Health and the control condition.