The representation of skin of colour in the Australasian Journal of Dermatology

Skin of colour (SoC) is underrepresented in dermatological literature with an associated reduction in clinician confidence diagnosing and managing skin conditions in patients with SoC.1-3 It is reported that half of Irish dermatology clinicians are not confident in diagnosing patients with SoC.1 Furthermore, confidence varies with the clinical context and a 2018 study found 85% of Australian dermatologists did not feel confident managing cosmetic skin issues in SoC.2 SoC representation in Australasian dermatology educational resources is largely undocumented. This study aimed to assess representation of SoC and associated demographic factors in published images in the Australasian Journal of Dermatology (AJD). We reviewed images in five online issues of AJD from February 2022 to February 2023. Skin type represented in images was categorised from I-VI according to Fitzpatrick skin type (FST), then grouped into 'light' (FST I–II), 'medium' (III–IV) and 'dark-skin' (V–VI) images. For each image, the following variables were recorded: dermatological condition, gender, article country of origin, age and ethnicity. Articles and captions were assessed for documentation of patient demographics and skin phototype. FST was assessed by the same author (LF) and with agreement from another author (HK) in cases where uncertainty arose. Histological, dermatoscopic, mucosal and duplicate images depicting the same patient were excluded. Images where both authors agreed image quality or framing meant skin phototype could not be determined were excluded. Two hundred and twenty-seven images were included in the analysis. The majority of images were FST II (36%) or III (33%) followed by FST IV (18%), FST I (7%) and FST V (6%) (Figure 1). No images showed FST VI skin. Only nine out of the 227 images (4%) had the FST skin type clearly stated at any point in the article or image caption. Overall, 43.6% and 50.7% of images were of light and medium skin, respectively (Figure 2). Dark skin was shown in 5.7% of images. Gender representation was evenly distributed; 37% of images identified females and 41% males. Gender was not stated for 23% of imaged patients. No imaged patients were stated as identifying as non-binary. There were no significant differences in representation of skin phototype by gender. Patient ethnicity was not mentioned for 78% of images. The most common listed ethnicity was Caucasian (Table 1). Indigenous Australasian ethnic groups including Māori, Aboriginal and Torres Strait Islander were not identified in any of the images. Our results are aligned with international analyses where SoC representation varies from 4%–18%.4, 5 Online resources may have greater SoC representation, varying from 17%–22% across examined websites6, 7. Interestingly, DermNet NZ was noted as having low SoC representation (2.8% in 2021)7 and editors have since focused on obtaining images representing dark skin types. Although far from equitable, skin of colour representation in dermatological literature is improving. A review spanning 1993–2021 found a 7.8-fold increase in the number of SoC articles published.8 Other studies have found representation varies by condition with lower SoC representation for skin cancers.6 In some resources, imaging of melanoma and basal cell carcinoma in dark skin is as low as 2.9% and 2.4%, respectively.7 Vitiligo and keloid scars are more likely to be imaged in SoC.6 Furthermore, in dermatology atlases and textbooks SoC images constitute a greater proportion of "stigmatising" conditions, such as sexually transmitted infections.9 The basis for this is not clear and there may be bias relating to conditions for which publishers use SoC images. Limitations of this study include determining FST from images without knowing burning or tanning tendencies. There was also potential for bias, as authors were unblinded to dermatological diagnoses. Lastly, skin type is a continuous variable and determining when skin becomes "skin of colour" is subjective. In conclusion, we found SoC images were under-represented in the AJD. Potential strategies for improvement include inviting authorship from countries with majority SoC population, inclusion of SoC targets in journal objectives and special issues with focus on SoC and indigenous skin health. Of course, equity likely requires overrepresentation of skin types that have been traditionally underrepresented, so as to provide increased educational resource to improve clinical confidence in managing patients with SoC. Finally, given most articles did not identify imaged patient demographics we recommend requesting image captions include basic patient data such as age, gender, ethnicity and FST. This would allow regular audits and facilitate accountability in goals towards equity for dermatology patients with SoC in Australasia. Open access publishing facilitated by The University of Auckland, as part of the Wiley - The University of Auckland agreement via the Council of Australian University Librarians. None. Harriet Kennedy is the co-editor in chief of the Australasian Journal of Dermatology.