Date of Award

January 2026

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Beverley J. Sheares

Abstract

Background: The ability to accurately recognize cutaneous manifestations of disease is a foundational clinical skill. However, current undergraduate medical education heavily relies on lighter skin tone prototypes, leaving trainees underprepared to diagnose conditions in patients with melanin-rich skin.Research Aims: This study aimed to evaluate the efficacy of a brief, targeted, visually comparative digital intervention by measuring its impact on medical students' immediate diagnostic accuracy and self-reported clinical confidence in identifying dermatologic conditions across a diverse range of skin tones. Hypothesis: We hypothesized that a brief, targeted digital educational intervention featuring side-by-side visual comparisons of dermatologic conditions across a range of skin tones would significantly improve medical students' immediate diagnostic accuracy and self-reported clinical confidence in recognizing dermatologic disease in Skin of Color (SoC). Methods: This single-arm, pre- and post-intervention study included 37 medical students across various training levels (M1–M4, M5, MD-PhD) at the Yale School of Medicine. Participants first completed a pre-survey assessing prior educational exposure, self-reported preparedness, and baseline diagnostic confidence. They then completed a pre-intervention diagnostic image test, followed by an image-based educational module featuring side-by-side visual comparisons of 18 dermatologic conditions across a range of skin tones to emphasize variation of presentation. Finally, participants completed a matched post-intervention diagnostic test and a post-survey to reassess diagnostic confidence and capture qualitative module evaluation. Results: At baseline, 94.6% of participants reported frequent educational exposure to findings in light skin tones, while 86.5% were rarely or never exposed to findings in the darkest skin tones. Consequently, 78.4% of the cohort felt underprepared to recognize dermatologic conditions across the full skin tone spectrum, a deficit that persisted even among final-year (M4) students. Following the digital intervention, mean diagnostic accuracy increased significantly from 56.5% to 90.4% (p < 0.001). Furthermore, diagnostic confidence for clinical findings presented in the darkest and medium skin tone tertiles demonstrated significant mean shifts of +1.16 and +0.73, respectively, on a 5-point Likert scale (both p < 0.001). Overall, 97.3% of participants agreed that the comparative module filled critical gaps in their medical education. Scientific impact and relevance for communities of interest: Medical students are systematically undertrained in recognizing dermatologic manifestations in SoC, a deficit that risks compromising care for patients with melanin-rich skin. Our findings demonstrate that a brief, comparative digital curriculum is associated with substantial, immediate improvements in diagnostic accuracy and clinical confidence. Moving forward, institutions should prioritize the formal, longitudinal integration of diverse visual images within the curriculum to achieve visual learning equity. Ultimately, this integration is critical to reducing diagnostic delays, minimizing disparities, and ensuring safe, equitable diagnostic care for all patient communities.

Comments

This is an Open Access Thesis.

Download
Open Access

This Article is Open Access

Share

COinS