이 페이지는 자동 번역되었으며 번역의 정확성을 보장하지 않습니다. 참조하십시오 영문판 원본 텍스트의 경우.

Effect of a Narrative Educational Podcast on Stereotypes and Prejudice Among Medical Students (LISTEN)

2026년 11월 13일 업데이트: Flamand-Roze Emmanuel, Sorbonne University

Learning to Improve Stereotypes Through Educational Narratives (LISTEN): A Randomized Controlled Trial

BACKGROUND Biases toward certain groups of patients (weight bias, ageism, racism, ableism and sexism) are common among health professionals and can affect the quality and equity of care. Traditional teaching formats such as lectures have difficulty engaging medical students on these deeply rooted attitudes. "Augusta's Oath" (original French title: "Le Serment d'Augusta") is a 16-episode narrative educational podcast (about 60 minutes per episode) produced by the Faculty of Health of Sorbonne University (Paris, France). Combining care stories, patient testimonies and insights from the humanities, it is used in an elective teaching unit to encourage reflection and change how future physicians approach patients. A previous uncontrolled study suggested that listening to the podcast prompted changes in perspective among medical students, but its specific effect on biases has never been tested in a controlled trial with validated measures.

OBJECTIVE To assess whether listening to targeted podcast episodes, combined with reflective writing, specifically reduces five types of bias: weight bias, ageism, racism, ableism and sexism.

METHODS This is a randomized controlled trial with two parallel groups, offered to medical students enrolled in the podcast teaching unit at Sorbonne University. About 500 students who agree to take part in the research are randomly assigned (1:1, balanced by sex and year of study) to one of two listening programmes of 8 episodes each (about 8 hours of listening), with a short written reflection after each episode.

  • Group A listens to episodes on weight bias, racism and sexual and gender-based violence, plus control episodes that do not address the biases measured.
  • Group B listens to episodes on disability and ageism, plus a different set of control episodes.

Each group therefore serves as the comparison group for the biases addressed only in the other group ("mirror" design). Participants commit not to listen to the other group's episodes until they have completed the study questionnaires.

At the end of the listening programme, participants complete attitude questionnaires once: AFA and ATOP (weight bias), ATIBI (recognition of implicit racial bias), AMMSA (acceptance of myths about sexual aggression), UCLA-GAS-F (attitudes toward older adults) and the revised ATPD (attitudes toward people with disabilities). Scores are compared between the two groups.

Participation in the research is voluntary and has no effect on course validation: students who decline follow the same teaching unit without any data collection. Data are pseudonymized, and participants who complete all questionnaires receive compensation for the time spent on research tasks.

EXPECTED RESULTS The investigators expect each group to show more favourable attitudes than the other group specifically on the biases addressed by its own episodes. Positive results would support the wider use of this narrative podcast as a teaching tool in other French-speaking medical schools.

연구 개요

상태

아직 모집하지 않음

정황

개입 / 치료

상세 설명

BACKGROUND Biases related to body weight, age, race or ethnicity, disability and gender are documented among health professionals and can contribute to inequities in care. Lecture-based teaching has limited impact on these deeply rooted attitudes. "Augusta's Oath" (original French title: "Le Serment d'Augusta") is a narrative educational podcast (16 episodes of about 60 minutes) produced by the Faculty of Health of Sorbonne University (Paris, France), combining care narratives, patient testimonies and insights from the humanities. A previous uncontrolled study of 478 medical students, based on text-mining of their written reflections, suggested that listening to the podcast prompted changes in perspective. However, no controlled study using validated attitude measures has yet assessed its specific effect on biases.

OBJECTIVE To assess whether listening to targeted podcast episodes, combined with reflective writing, specifically reduces five types of bias: weight bias, racism, sexism, ageism and ableism.

DESIGN Single-centre, prospective, two-arm, parallel-group randomized controlled trial with a mirror design. Each group listens to a distinct set of 8 episodes, including target episodes addressing specific biases and control episodes addressing other medical humanities topics. Each group therefore serves as the comparator for the biases targeted in the other group. The trial compares two active interventions and estimates a differential effect (Group A minus Group B), not the effect of each pathway versus no intervention.

  • Group A: target episodes on weight bias (episode 1), racism (episode 10) and sexual and gender-based violence (episode 11); control episodes 2, 3, 4, 5 and 9.
  • Group B: target episodes on disability (episodes 14 and 15) and ageism (episode 16); control episodes 6, 7, 8, 12 and 13.

Episodes were allocated after independent rating by two investigators so that Group B, the comparator for sexism outcomes, contains no episode with a predominantly anti-sexist narrative focus.

PARTICIPANTS AND SETTING 2nd- to 6th-year medical students at the Faculty of Health, Sorbonne University, enrolled in the "Augusta's Oath" teaching unit, who give informed consent to the research. Target sample: about 500 participants. Research participation is voluntary and independent of course validation: students who decline follow the same teaching unit (8 episodes and 8 written comments) without any data collection.

RANDOMIZATION AND ALLOCATION Computer-generated 1:1 allocation (R, predefined seed), stratified by sex and year of study, using permuted blocks of variable size (4 or 6) within strata. Randomization is performed by the study statistician on a pseudonymized table containing only a technical identifier, sex and year of study. An independent administrator, who alone holds the link between identities and identifiers, configures conditional access on the university Moodle platform so that each participant can only access the 8 episodes of their group. Participants commit not to listen to the other group's episodes until they have completed the study questionnaires.

MASKING Participants know the topics studied and the titles of their episodes, but are not informed of the group-specific hypotheses. Outcomes are self-reported online without any assessor. Before transfer of the analysis database, the independent administrator assigns new random identifiers (distinct from those used for randomization), codes the groups neutrally and provides adherence variables in derived form (yes/no). The correspondence between codes and groups is revealed only after database lock and completion of the primary analyses.

INTERVENTION Individual listening on Moodle to the 8 assigned episodes, each followed by a written reflective comment of at least 100 words (about 12 hours in total), as part of the teaching unit requirements. The research questionnaires become accessible once all 8 comments have been submitted.

OUTCOMES

Self-report scales completed once, at the end of the listening programme (by 15 July 2027). Scores used in the primary outcome:

  • Weight bias: Anti-Fat Attitudes Questionnaire (AFA; French adapted version; mean of 13 items, 0-8)
  • Racism: Attitudes Toward Implicit Bias Instrument (ATIBI; 25 published items administered, primary score = sum of the 18 items of the validated version, 18-108)
  • Sexism: Acceptance of Modern Myths about Sexual Aggression scale (AMMSA; validated French version; mean of 30 items, 1-7)
  • Ageism: UCLA Geriatrics Attitudes Scale, French version (UCLA-GAS-F; sum of 14 items, 14-70)
  • Ableism: revised Attitudes Toward People With Disabilities scale (ATPD; mean of 20 items, 1-5) Secondary weight-bias measure: Attitudes Toward Obese Persons scale (ATOP; 0-120).

ATOP, ATIBI and ATPD were translated into French by the study team. Self-reported adherence, prior listening to the podcast, and sociodemographic and personal characteristics are also collected.

STATISTICAL ANALYSIS The primary analysis follows the intention-to-treat principle: all randomized participants are analysed in their allocated group, regardless of actual listening or questionnaire completion, with multiple imputation (MICE) of missing outcome scores. Scores are harmonized (higher = more bias) and standardized. The primary outcome is a global test of the mean aligned standardized between-group effect across the five dimensions (AFA, ATIBI, AMMSA, UCLA-GAS-F, ATPD), estimated with a multivariate linear model adjusted for the stratification variables (sex and year of study), two-sided alpha 5%. If the global test is significant, each dimension is tested by linear regression adjusted for sex and year of study with robust (HC3) standard errors, with Holm correction; a dimension is confirmed if its adjusted p-value is < 0.05 and the effect is in the hypothesized direction. A minimal relevant difference of 0.25 standard deviations is set a priori.

ATOP is analysed as a secondary outcome, and the global test is repeated with ATOP replacing AFA as a sensitivity analysis. Other sensitivity analyses include unadjusted and extended models, complete-case analysis, missing-not-at-random scenarios (delta adjustment and tipping-point analysis), non-parametric tests, and analyses restricted to participants who completed the course pathway or reported full adherence.

Exploratory analyses, with Benjamini-Hochberg correction, include effect moderators (study phase: 2nd-3rd vs 4th-6th year; prior listening; gender), correlations between bias dimensions within groups, common structure and bias profiles, and psychometric properties and measurement invariance of the instruments.

Power: with 250 participants per group, each uncorrected test has 80% power for d = 0.25; assuming an inter-outcome correlation of 0.3, the global test has about 72% power for a mean effect of d = 0.15 and 92% for d = 0.20.

ETHICS AND DATA PROTECTION Approved by the Sorbonne Université Research Ethics Committee (IRB00014283), protocol no. CER-2026-FLAMAND-ROZE-LISTEN-00685, on 14 September 2026. Data are pseudonymized and processed in compliance with the EU General Data Protection Regulation. Participants who complete all questionnaires receive a 40-euro compensation for the time spent on research tasks.

PREREGISTRATION The full study protocol and statistical analysis plan (version 1.0) are preregistered on the Open Science Framework.

연구 유형

중재적

등록 (추정된)

500

단계

  • 해당 없음

연락처 및 위치

이 섹션에서는 연구를 수행하는 사람들의 연락처 정보와 이 연구가 수행되는 장소에 대한 정보를 제공합니다.

연구 연락처

연구 장소

참여기준

연구원은 적격성 기준이라는 특정 설명에 맞는 사람을 찾습니다. 이러한 기준의 몇 가지 예는 개인의 일반적인 건강 상태 또는 이전 치료입니다.

자격 기준

공부할 수 있는 나이

18년 이상 (성인, 고령자)

건강한 자원 봉사자를 받아들입니다

예

연구 대상 성별

모두

설명

Inclusion Criteria:

  • Medical students in their 2nd to 6th year at the Faculty of Health, Sorbonne University
  • Enrolled in the "Le Serment d'Augusta" teaching unit
  • Informed consent to participate in the research

Exclusion Criteria:

  • Refusal to participate in the research or objection to the use of their data (these students follow the same teaching unit without any data collection)

공부 계획

이 섹션에서는 연구 설계 방법과 연구가 측정하는 내용을 포함하여 연구 계획에 대한 세부 정보를 제공합니다.

연구는 어떻게 설계됩니까?

디자인 세부사항

  • 주 목적: 다른
  • 할당: 무작위
  • 중재 모델: 병렬 할당
  • 마스킹: 없음(오픈 라벨)

무기와 개입

참가자 그룹 / 팔
개입 / 치료
실험적: Group A: Podcast Episodes on Weight Bias, Racism and Sexism

Listening to 8 episodes of the educational podcast "Augusta's Oath" (about 1 hour each), with a written reflective comment after each episode.

Target episodes: 1 (weight bias), 10 (racism) and 11 (sexual and gender-based violence).

Control episodes (addressing other medical humanities topics, not the biases measured): 2, 3, 4, 5 and 9.

Serves as the comparator for disability and ageism outcomes.

Individual listening on the university Moodle platform to the 8 assigned episodes of group A, each followed by a written reflective comment (about 12 hours in total), as part of the teaching unit requirements.

다른 이름들:
  • Augusta's Oath
실험적: Group B: Podcast Episodes on Disability and Ageism

Listening to 8 episodes of the educational podcast "Augusta's Oath" (about 1 hour each), with a written reflective comment after each episode.

Target episodes: 14 and 15 (disability) and 16 (ageism). Control episodes (addressing other medical humanities topics, not the biases measured): 6, 7, 8, 12 and 13.

Serves as the comparator for weight bias, racism and sexism outcomes.

Individual listening on the university Moodle platform to the 8 assigned episodes of group B, each followed by a written reflective comment (about 12 hours in total), as part of the teaching unit requirements.

연구는 무엇을 측정합니까?

주요 결과 측정

결과 측정
측정값 설명
기간
Composite Bias Score: Global Aligned Standardized Between-Group Effect Across Five Bias Scales
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Single pre-specified composite outcome: mean of the five standardized between-group differences (Group A minus Group B), each aligned with the hypothesized direction, expressed in standard deviation units (Cohen's d). The five component scores are: Anti-Fat Attitudes Questionnaire (AFA), Attitudes Toward Implicit Bias Instrument (ATIBI, 18-item score) and Acceptance of Modern Myths about Sexual Aggression (AMMSA), expected to favour Group A; UCLA Geriatrics Attitudes Scale, French version (UCLA-GAS-F) and revised Attitudes Toward People With Disabilities scale (ATPD), expected to favour Group B. Positive values indicate an effect in the hypothesized direction. Estimated with a multivariate linear model adjusted for sex and year of study. Each component scale is also reported as a separate secondary outcome.

Up to 8 months after enrollment (once, at the end of the listening programme)

2차 결과 측정

결과 측정
측정값 설명
기간
Weight bias: Anti-Fat Attitudes Questionnaire (AFA) score
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Mean of 13 items (French adapted version), range 0-8; higher scores indicate stronger anti-fat attitudes. Between-group difference adjusted for sex and year of study.

Up to 8 months after enrollment (once, at the end of the listening programme)
Recognition of implicit racial bias: Attitudes Toward Implicit Bias Instrument (ATIBI) score
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Sum of the 18 items of the validated version (items 12 and 16 reverse-scored), range 18-108; higher scores indicate more favourable attitudes toward recognizing and managing implicit bias. Between-group difference adjusted for sex and year of study.

Up to 8 months after enrollment (once, at the end of the listening programme)
Sexism: Acceptance of Modern Myths about Sexual Aggression (AMMSA) score
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Mean of 30 items (validated French version), range 1-7; higher scores indicate stronger acceptance of myths about sexual aggression. Between-group difference adjusted for sex and year of study.

Up to 8 months after enrollment (once, at the end of the listening programme)
Ageism: UCLA Geriatrics Attitudes Scale, French version (UCLA-GAS-F) score
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Sum of 14 items (9 reverse-scored), range 14-70; higher scores indicate more favourable attitudes toward older patients. Between-group difference adjusted for sex and year of study.

Up to 8 months after enrollment (once, at the end of the listening programme)
Ableism: revised Attitudes Toward People With Disabilities (ATPD) score
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Mean of 20 items (6 reverse-scored), range 1-5; higher scores indicate more stigmatizing attitudes toward people with disabilities. Between-group difference adjusted for sex and year of study.

Up to 8 months after enrollment (once, at the end of the listening programme)
Weight bias: Attitudes Toward Obese Persons (ATOP) score
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

Algebraic sum of 20 items plus a constant of 60, range 0-120; higher scores indicate more positive attitudes toward people with obesity. Between-group difference adjusted for sex and year of study. Also used in a sensitivity analysis replacing AFA in the global test.

Up to 8 months after enrollment (once, at the end of the listening programme)

기타 결과 측정

결과 측정
측정값 설명
기간
AFA subscores (Dislike, Fear of fat, Willpower; means, range 0-8)
기간: Up to 8 months after enrollment (once, at the end of the listening programme)

ATIBI 25-item exploratory score (sum, range 25-150)

Up to 8 months after enrollment (once, at the end of the listening programme)

공동 작업자 및 조사자

여기에서 이 연구와 관련된 사람과 조직을 찾을 수 있습니다.

스폰서

수사관

  • 수석 연구원: Prof Emmanuel Flamand-Roze, M.D. Ph.D., Sorbonne University

연구 기록 날짜

이 날짜는 ClinicalTrials.gov에 대한 연구 기록 및 요약 결과 제출의 진행 상황을 추적합니다. 연구 기록 및 보고된 결과는 공개 웹사이트에 게시되기 전에 특정 품질 관리 기준을 충족하는지 확인하기 위해 국립 의학 도서관(NLM)에서 검토합니다.

연구 주요 날짜

연구 시작 (추정된)

2026년 11월 16일

기본 완료 (추정된)

2027년 7월 15일

연구 완료 (추정된)

2027년 12월 31일

연구 등록 날짜

최초 제출

2026년 11월 9일

QC 기준을 충족하는 최초 제출

2026년 11월 13일

처음 게시됨 (실제)

2026년 11월 19일

연구 기록 업데이트

마지막 업데이트 게시됨 (실제)

2026년 11월 19일

QC 기준을 충족하는 마지막 업데이트 제출

2026년 11월 13일

마지막으로 확인됨

2026년 11월 1일

추가 정보

이 연구와 관련된 용어

기타 연구 ID 번호

  • CER-2026-LISTEN-00685
  • CER-2026-FLAMAND-ROZE-00685 (기타 식별자: Sorbonne Université Research Ethics Committee)

개별 참가자 데이터(IPD) 계획

개별 참가자 데이터(IPD)를 공유할 계획입니까?

아니요

IPD 계획 설명

Individual participant data will not be shared outside the research team. The data include sensitive personal information (e.g., family origin, exposure to sexual or gender-based violence, disability, obesity) collected from medical students of a single faculty, which creates a risk of re-identification even after de-identification. In addition, the information provided to participants and the ethics approval do not cover sharing of individual data with external researchers, and data are processed under the EU General Data Protection Regulation. Aggregated results and the statistical analysis code will be made available with the study publications.

약물 및 장치 정보, 연구 문서

미국 FDA 규제 의약품 연구

아니

미국 FDA 규제 기기 제품 연구

아니

이 정보는 변경 없이 clinicaltrials.gov 웹사이트에서 직접 가져온 것입니다. 귀하의 연구 세부 정보를 변경, 제거 또는 업데이트하도록 요청하는 경우 register@clinicaltrials.gov. 문의하십시오. 변경 사항이 clinicaltrials.gov에 구현되는 즉시 저희 웹사이트에도 자동으로 업데이트됩니다. .