Is there an under-representation of skin of colour images during the COVID-19 outbreak?

Dear Editor,

The lack of representation of skin of colour during COVID-19 pandemic in dermatologic literature has been recently discussed [1]. According to some authors [1], there might be a discrepancy between the increased incidence of COVID-19 among dark-skinned patients and the under-representation of their skin manifestations in published series [1]. The previous letter found a non-negligible echo on social medias as illustrated by Altmetric results (https://wiley.altmetric.com/details/83085628). For instance, the account Brown Skin Matters, which is involved in training for non-white skin dermatology, advertised on Twitter (12,7 K followers) and Instagram (36,6 K followers) about this paper stating that “Brown skin [is] almost totally absent in publications about COVID-19 dermatological symptoms” and made a parallel with the Black lives matter movement, giving a political flavor to the article. As academic dermatologists involved in educational training, we agree on the necessity of spreading and teaching the specificities of dark-skinned diseases. Lester et al warned previously against the under-representation of images of skin of colour in dermatology textbooks and set photos [2]. However, this issue cannot be extrapolated to publications related to COVID-19 skin manifestations.

The overall prevalence of cutaneous manifestations in COVID-19 confirmed patients – regardless of ethnicity – is not clearly known and varies widely from 0.2 to 20% depending on the series [3]. An increased prevalence of infection among dark-skinned individuals would not necessarily imply increased prevalence of skin manifestations. In an international registry of 682 COVID-19 patients with dermatologic conditions [4], Hispanic and Afro-American patients constituted only 5.0% and 1.9% of cases. Pangti et al. found that vascular cutaneous manifestations were uncommon in Indian patients with phototypes IV and V [5].

Lester et al. remind us that 30% of the COVID-19 patients in the UK and US are dark-skinned, but this prevalence is not necessarily generalizable to other European countries, which were hit by the COVID-19 outbreak, and where most skin manifestations have been described so far. Indeed, the inclusion period of their latest study (between December 31, 2019 and May 3, 2020) comprises mostly series from Spain, Italy and France and very few published reports from the US or the UK [3]. While ethnical background is a standard demographic data in the US scientific literature, ethnic origins collecting policies in official statistics vary according to countries within the EU [5]. Germany, France, Italy or Spain do not collect such data [5]. This probably accounts for the paucity of articles reporting ethnic information in COVID-19 dermatology reports published so far (only 6 articles out of 36 in their review). Furthermore, the standard distinction of ethnic groups, as done in the US literature, has not the same relevance in Europe: for instance, the question of differentiating “Hispanic” patients from other patients from the Mediterranean area, such as Portuguese or Italians for instance is highly debatable as patients share similar skin phototype.

Although we do not refute that skin of color may be under-represented in dermatology teaching, it cannot be stated that European series under-reported clinical images from dark-skinned patients on purpose or unintentionally. Dark-skinned individuals may be less likely to present cutaneous manifestations [4], [5]. Absence of collection of ethnic data is related to institutional standards and challenged by the complexity of population mixing in Europe [6].

Specific studies of cutaneous manifestations among dark-skinned patients with COVID-19 is of interest in countries with high incidence of COVID-19 such the US or Brazil.

Funding sources

None.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.