Images of dark skin in top emergency medicine journals: A cross‐sectional analysis of images of emergent cutaneous disorders. (29th September 2021)

• Record Type: Journal Article
• Authors: Bell, Geovonni; Holmes, Sherita; Gillespie, Scott; Wood, Anna; Murray, Brittany L.
• Is Part Of: AEM education and training. Volume 5(2021)supplement 1
• Page Start: S76
• Page End: S81
• Publication Date: 2021-09-29
• Subjects: Emergency medicine -- Study and teaching -- Periodicals; Emergency medicine -- Study and teaching -- United States -- Periodicals; Periodicals; 616.025
• DOI: 10.1002/aet2.10683
• Languages: English
• ISSNs: 2472-5390

Abstract

Abstract: Introduction: Emergency medicine (EM) physicians must recognize emergent cutaneous disorders (CDs) in patients of all skin tones. In other medical specialties, images of CDs in light‐skinned individuals (LSI) are published more frequently than images of CDs in dark‐skinned individuals (DSI). This study aims to determine the representation of LSI versus DSI in images of emergent CDs published in top EM journals. Methods: This is a cross‐sectional analysis of CD images published from 2015 to 2020 in the six most influential EM journals as determined by Eigenfactor. The 2016 Model of the Clinical Practice of Emergency Medicine (EM Model) by the American Board of Emergency Medicine was used to classify CDs as "emergent, " "nonemergent, " or "not listed." The Fitzpatrick skin tone scale was used to classify skin tone as light, dark, or indeterminate. Two blinded reviewers classified each image; for disagreements, a third blinded reviewer determined the final classification. Descriptive statistics and chi‐square were used to analyze the data. A kappa coefficient was used to determine reviewer agreement (LSI vs. DSI), and a weighted kappa coefficient was used for agreement between individual Fitzpatrick categories. Results: There were 314 images of CDs. Forty images were indeterminate, and one image was excluded, leaving 273. Of the 273 images analyzed, 44.0% were emergent, 8.0% were nonemergent, and 48.0% were not listed in the EM Model. DSI comprised 13.6% of images. For emergent CDs, 85.0% were LSI versus 15.0% DSI. For nonemergent CDs, 27.3% were DSI, and for CDs not listed in the EM Model, 9.9% were DSI. The kappa coefficient for reviewer agreement between LSI and DSI was 0.76 (95% confidence interval [CI] = 0.65 to 0.87) and the weighted kappa coefficient for agreement between Fitzpatrick categories was 0.70 (95% CI = 0.64 to 0.76), showing substantial agreement. Conclusion: For emergent and nonemergent CDs, images of LSI were published more than those of DSI in top EM journals.