Evaluation of a Digital Clinician-Training Module for Treatment of Transgender and Gender Diverse Patients With Eating Disorders: Pre-Post Study

Introduction

Among transgender and gender diverse (TGD) people, 18% were diagnosed with an eating disorder (ED) in the past year, compared to <2% of cisgender people [1]. TGD people experience internal stressors, such as gender dysphoria, and external stressors, such as discrimination, which contribute to an elevated rate of EDs.

Clinicians may receive insufficient medical training about EDs [2]. Few clinicians are specifically educated on how to care for TGD people with EDs, and thus they may fail to address the unique challenges faced by these patients, hindering their recovery [3]. Furthermore, clinicians report variable attitudes toward working with TGD patients, which may be improved through convenient and accessible targeted digital training [4]. However, no in-depth clinician training activities currently focus on TGD people with EDs [5].

In this study, we report on the development and evaluation of a training module on treating TGD people with EDs for health care workers in primary care settings (Multimedia Appendix 1). The structure of the module aligns with principles of adult learning theory (eg, andragogy) and student-centered learning (eg, scaffolding) by emphasizing self-directed, practice-based learning and application of progressively more advanced skills to clinical scenarios [6].

As lack of institutional support represents a significant obstacle to producing clinician-training materials for marginalized populations, we demonstrate the feasibility of a digital approach requiring minimal technical and financial support [6].

Methods

Module Structure

Developed by the authors, the module is intended to be completed independently in approximately one hour (Multimedia Appendix 1). The learner advances through 6 sections in the slide deck, including an overview, instructional slides, and ungraded, interactive knowledge checks, where learners can click the question slide to reveal the answer. The module is hosted on Google Slides, allowing the learner to skip or navigate sections in any order.

Module Evaluation

Participants were Harvard Medical School medical students recruited using a school-wide mailing list. Students completed a preintervention survey, the module, and a postintervention survey via email.

In the preintervention and postintervention surveys, participants rated their agreement with a set of statements designed based on prior studies [7-9] that evaluated beliefs (preintervention: α=.87; postintervention: α=.88), attitudes (preintervention: α=.88; postintervention: α=.93), and self-efficacy (preintervention: α=.87; postintervention: α=.85) toward working with TGD patients with EDs on a 7-point Likert-scale from “strongly disagree” (1) to “strongly agree” (7). The survey items are shown in the Results section. Sociodemographic details and participant satisfaction were also collected [7,10]. Participants were required to complete all questions.

Using R (version 4.4.1; R Project for Statistical Computing) in Rstudio, mean and SD for the survey data were calculated; mean differences and the significance level between paired questions were determined with paired t tests, a method that is robust for n>30.

Ethical Considerations

The Harvard-Longwood Campus Institutional Review Board approved this study (IRB24-0636). Informed consent was collected. All data were deidentified, and participants received a US $50 e-gift card.

Results

Of 102 participants who completed the preintervention survey, 87 (85%) also completed the postintervention survey and were included in the study. Most participants were cisgender, white or Asian, and had TGD and/or lesbian, gay, bisexual, transgender, queer/questioning, intersex, and asexual (LGBTQIA+) friends (Table 1).

As shown in Table 2, paired 2-tailed t tests identified significant increases in all 3 subsections; the largest increases were observed on self-efficacy statements, such as identifying harm reduction strategies for gender dysphoria (+2.39; P<.001). Only 2 statements, on believing TGD identities are valid and being interested in learning about TGD health interventions, did not significantly increase, potentially due to high pretest scores. Participants indicated high levels of satisfaction, with mean ratings between 5.70 and 6.45.

The module was improved based on feedback (eg, adding more case study questions); the module will be made eligible for continuing medical education credits and promoted through the National LGBTQIA+ Health Education Center.

Discussion

To address disparities in ED risk and outcomes among TGD patients, we developed an hour-long self-paced learning module that included 6 sections with accompanying review questions after each section, and we piloted the module with a cohort of 87 medical school students. Prior to completing the module, students scored highly in the attitudes and beliefs subsections but lower in the self-efficacy subsection, suggesting that while the vast majority of students supported TGD people and had TGD and/or LGBTQIA+ friends, they remained uncertain about how to help TGD patients in a clinical setting. After completing the education module, the students showed statistically significant improvements in all 3 subsections, with the largest improvements in self-efficacy. Students reflected that they gained valuable information from the module, oftentimes despite already having some familiarity with the topic, and they felt confident applying this new information in a clinical setting.

Consistent with prior studies and adult learning theory [6], students found application-based questions to be most helpful. In particular, students reported that the case study questions, which present hypothetical patient care scenarios, meaningfully improved their self-efficacy toward applying concepts in the clinical setting.

Results are limited by the student participant pool, who may not be representative of primary care workers, especially those from different locations and/or political climates. Social desirability bias may have skewed the scores, and skills-based learning was not tested. Participant interviews may further elucidate module strengths and weaknesses; the module may also be enhanced through interactive elements not possible in Google Slides, such as clickable answers.

Supportive and respectful attitudes alone are insufficient to ensure that TGD patients receive high-quality medical care. During a time of limited institutional support for LGBTQIA+ health equity, publicly and digitally accessible clinician training activities that apply this free, simple, and effective model can allow students to learn more advanced clinical skills for addressing a specific health disparity in a subpopulation of LGBTQIA+ patients.