Skin of color
Rosacea on darker skin: why it's missed
Rosacea is reported far less often in people with darker skin, and the dermatology literature is clear that a meaningful part of that gap is not biology, it is under-recognition. The redness rosacea is screened for is harder to see on pigmented skin, so the disease is missed, and the image tools meant to help inherit the same blind spot. Here is the evidence, and what actually signals the disease across skin tones. This is informational, not medical advice.
Under-diagnosis, not absence
Reported rosacea prevalence in Fitzpatrick IV and V skin sits at roughly 6 to 8 percent of the diagnosed rosacea population, a number that does not match what dermatologists who work with darker-skinned patients see in clinic. The skin-of-color review literature (Alexis 2019; Maliyar and Abdulla 2022) is explicit about why: a meaningful share of the apparent gap is diagnostic masking. The features rosacea is most commonly screened on are visual and contrast-dependent, and on pigmented skin they are harder to see.
Telangiectasias are described in the Maliyar review as "often not observed in highly pigmented skin" with the naked eye. The vessels are there; the contrast against the background simply is not enough to make them obvious. The same is true of persistent erythema. So the disease is present at rates closer to what general rosacea epidemiology predicts, but the screening apparatus is calibrated against features that do not show up the same way, and it undercounts.
The automated tools make it worse, and there's data
Dermatology image-model sensitivity, lighter vs darker skin
Sensitivity (higher is better) of two widely-deployed dermatology image-classification models, tested on the Diverse Dermatology Images set. Source: Daneshjou et al., Science Advances 2022 (eabq6147).
In 2018, Adamson and Smith published a viewpoint in JAMA Dermatology (2018;154(11):1247-1248) warning that machine-learning systems for dermatology were being trained on image datasets that under-represented darker skin types, and would therefore perform worse on them. In 2022, the empirical follow-up arrived: Daneshjou and colleagues (Science Advances 2022, eabq6147) built the Diverse Dermatology Images set, biopsy-confirmed images labeled by skin type, and ran two widely-deployed dermatology models against it. The drop was large, as the chart shows: sensitivity roughly a third of its lighter-skin value on darker skin. A redness-detecting tool built on the same kind of data will miss the same patients a human screener calibrated on light-skin erythema does.
What to track instead
The implication for anyone with darker skin tracking rosacea is direct: do not let visible redness be the only headline metric. The signal that travels better across skin tones is the sensory diary, how often and how intensely you burn, sting, and flush, alongside warmth and bumps. Photos still matter for your dermatologist, and a consistent daily image is still worth keeping, but the redness-only view is exactly the one that undercounts the disease on pigmented skin. Track the experience, not just the color.
Built to work across skin tones, not just the visible-redness ones.
Skinframe records the sensory experience, burning, stinging, flushing, warmth, alongside the daily photo, so the disease is tracked on the dimension that matches it for you, and your photos display on a neutral gray surface that is never tinted. Your photos stay in your own iCloud. One-time $29.99, or $4.99 a month with one-tap cancel.
Get Skinframe on the App StoreQuestions
Can you get rosacea on dark skin?
Yes. Rosacea occurs across all skin tones. It is reported less often in people with darker skin, but the dermatology literature attributes a meaningful part of that gap to under-recognition rather than true rarity: the persistent redness and visible vessels that rosacea is usually screened for are harder to see against more pigmented skin, so the disease is missed. If you have flushing, burning, stinging, or bumps, those are worth raising with a dermatologist regardless of skin tone.
Why is rosacea missed on darker skin?
Because the screening cues are visual and contrast-dependent. Telangiectasia (visible blood vessels) and erythema (redness) are described in the skin-of-color literature as often not observed with the naked eye on highly pigmented skin: the vessels and redness are present, but the contrast against the background is too low. Automated image-analysis tools inherit the same blind spot, because they were trained mostly on lighter skin. The signal that travels better across skin tones is the sensory experience: how often and how intensely you burn, sting, or flush.
What should I track for rosacea if I have darker skin?
Track the sensory diary alongside the photos: frequency and intensity of burning, stinging, and flushing, plus warmth and bumps, not just visible redness. A tool that surfaces only photo-derived redness as its headline metric is asking the wrong question of darker skin. Photos still help your dermatologist, but the redness-only view undercounts the disease on pigmented skin.
Read next: the full analysis in rosacea on every skin tone, or how to document rosacea for your dermatologist.
Sources: Daneshjou R, et al. Disparities in dermatology AI performance on a diverse, curated clinical image set. Science Advances 2022, eabq6147. Adamson AS, Smith A. Machine Learning and Health Care Disparities in Dermatology. JAMA Dermatology 2018;154(11):1247-1248. Maliyar K, Abdulla SJ, skin-of-color rosacea review 2022; Alexis et al. 2019. This page is informational and not medical advice.