Twenty-Two Patients in Fourteen Years

Some absences in the literature are gaps in disease. Most are gaps in counting. From the outside the two look identical, and the difference decides whether a population is healthy or merely unrecorded.

Melanoma that only exists if you link the records

Non-Hispanic American Indian and Alaska Native people have the second-highest invasive cutaneous melanoma incidence in the United States, after non-Hispanic white people. Overall rate 10.7 per 100,000, rising to 24.2 above age 55 and 34.5 in men over 55.

That figure — 2,151 incident melanomas across twenty years — was only obtainable by linking cancer registry data to Indian Health Service administrative records, explicitly to reduce racial misclassification.

Without the link, the cases are miscoded and the rate disappears. Which means the received impression that this is a low-risk population was partly a data-infrastructure effect.

And the trend is going the wrong way: incidence rising 2.5 percent a year in women, and 2.5 percent a year in regional and distant-stage tumors.

What that population has by way of dermatology is best captured by a survey administered at two community powwows in Denver. Of 225 respondents reporting at least one skin disease, 20.9 percent had ever seen a dermatologist. Forty-two percent were not interested in teledermatology at either home or clinic.

A convenience sample at a powwow is among the largest primary datasets in this area. The field’s own framing, from a 2016 review, is its opening sentence: dermatology is greatly understudied in the American Indian and Alaska Native population.

Australia counts the scabies and stops counting at the skin cancer

Australia runs the most developed skin cancer prevention research program in the world. It also contains the highest impetigo prevalence ever recorded anywhere — 49.0 percent, in Aboriginal communities, from a systematic review of 48 population-based studies.

Scabies prevalence in Aboriginal and Torres Strait Islander children in remote communities is estimated as high as 33 percent, described as the third highest in the world, against an endemic-country baseline of 5 to 10 percent in children. An earlier study from northern Australia put scabies in children at up to 50 percent, with up to 70 percent of children having skin sores.

An ivermectin mass drug administration study recorded that 69 percent of infants were infected in the first year of life.

An audit of two remote primary healthcare centers found skin conditions were the most common reason for presentation — 22 percent of all presentations — and that 75 percent of residents presented at least once with a skin infection. Its opening line: the burden of skin infections across all age groups in remote Australian Indigenous communities is currently unknown.

Head lice mean prevalence across nine Kimberley communities over ten timepoints: 48.4 percent, with repeat detection in two thirds of affected children. That paper also notes the rate was previously unknown.

A narrative review of health disparities in the Northern Territory puts skin infections at fifty-fold greater than in non-Indigenous Territorians. Rheumatic heart disease — which follows from streptococcal skin infection — at 260-fold.

Now the other half of the ledger.

The total published record of cutaneous malignancies in urban Indigenous Australians is one retrospective case series: 22 patients, two Sydney centers, fourteen years. Its own opening states that despite 3 percent of Australians identifying as Indigenous, cutaneous malignancies in these patients — incidence, risk factors and outcomes — have not been investigated.

Of those 22, 68.2 percent presented at stage II or higher.

A 2014 review of non-infectious skin disease in Indigenous Australians opens: the burden has not been previously examined, and there is a paucity of literature on the topic. What it could assemble suggested lower psoriasis and lower skin cancer, and higher lupus erythematosus and vitamin D deficiency.

Lower skin cancer. Which is precisely why the research money went somewhere else, in the country that has more skin cancer research than anywhere on earth.

And a systematic review of teledermatology satisfaction concluded, in plain words: no published research currently exists on the use of teledermatology among Australia’s Indigenous population. A systematic review returning a null set is the strongest possible way to document an absence.

The consequence is not hypothetical. As of April 2026, the Northern Territory was in the middle of a diphtheria outbreak — 131 cases, 97 of them cutaneous — after decades with no local toxigenic transmission. The published attribution is crowding and consequent poor skin health.

Asian or Pacific Islander, as a single box

There is essentially no dermatology literature specific to Native Hawaiian and Pacific Islander people. I searched PubMed on 13 September 2026 for Native Hawaiian and Pacific Islander combined in turn with melanoma, skin cancer and dermatology. Seventy-one records came back, and the overwhelming majority mention the group only as an unanalyzed stratum.

The reason to believe that matters is that the disaggregation has been done in other diseases, and it changes the answer every time.

A health workforce analysis found that Asian American, Native Hawaiian and Pacific Islander people are well represented in the health workforce as a collective — and that Native Hawaiian and Pacific Islander populations specifically, along with several Southeast Asian populations, are underrepresented. The category encompasses more than 50 ethnicities and 100 languages and is, as the authors put it, often treated as a monolith.

That is the cleanest published proof that aggregation manufactures a false appearance of parity. An aggregate that looks fine, hiding a subgroup that is not.

The same structure appears in breast cancer incidence, in heart failure incidence, and in a study of fatal police violence. In every case, the aggregate is uninformative and the subgroups vary widely.

In dermatology it has simply not been done. And the aggregation is not historical — a 2024 US Cancer Statistics analysis covering 96 percent of the US population reports rates for “non-Hispanic Asian or Pacific Islander” as a single combined category.

Even the field’s own summary carries the problem. A systematic review of skin cancer in people of color screened 2,666 publications and included 251 — for African, Asian, Hispanic or Latino, Pacific Islander and Native Indian descent combined, across thirty years. Two hundred and fifty-one studies for five continental-scale groupings. And the review reproduces, in its own category structure, the aggregation it is documenting.

Roma, and a word that means two things

I searched PubMed on 13 September 2026 for Roma as a population descriptor combined with a dermatological term. Nothing came back at all.

Broadened the same day — Roma, Romani, Gypsy and Travelers, against skin, dermatology and scabies — it returned 246 records which are almost entirely about travel medicine. Rickettsioses in tourists. Cutaneous larva migrans in returning holidaymakers.

The word Traveler means an ethnic group and it means a person on a trip, and medicine indexes the second one. A researcher actively looking for this literature would struggle to find it, which is a discoverability barrier on top of the absence.

The mechanism of the absence is stated plainly in a review of Irish Roma health: up until 2021, the national census in Ireland did not include Roma as an ethnicity category.

No census category means no denominator. No denominator means no incidence rate can ever be calculated, for anything, ever. That is not an oversight in dermatology. It is arithmetic that was decided elsewhere.

The scale of the thinness: a mixed-methods systematic review of discrimination against childbearing Romani women in European maternity care screened 900 papers and found three eligible qualitative studies. For the largest ethnic minority in Europe.

There is no scabies, impetigo, atopic dermatitis, psoriasis or skin cancer prevalence figure specific to Roma populations in the indexed peer-reviewed literature. If one is circulating, it did not come from there.

And the picture the training runs on

An audit of five major rheumatology educational resources coded 1,604 images: 86 percent light skin, 9 percent dark skin, 5 percent indeterminate.

Dermatology textbooks most commonly depict Fitzpatrick II and III and least commonly V and VI, differing significantly from the national skin type distribution at p below 0.001 for every textbook examined.

Across 6,719 posts and 2,547 unique patient images on plastic surgery journal and professional organization Instagram accounts, 88.14 percent of patient images showed white skin.

A 2026 editorial in the International Journal of Dermatology states that skin type classification still lacks a universally accepted framework, that skin of color is imprecise and often used simply to denote people who are not of Caucasian white origin, and flags the high rate of misdiagnosis that follows from atypical presentations and limited training.

Which closes the loop. The categories are imprecise, the images do not show the range, the registries collapse the groups, the census in one country did not list them, and the systematic reviews come back empty. Each of those is a small administrative decision. Together they are the reason a clinician can finish training having never seen a photograph of the thing they will be asked to recognize.

I cannot tell you what the evidence says about skin disease in Roma communities, or in Pacific Islander communities specifically, or in most Indigenous populations outside northern Australia, because nobody has written it down. What I can do is say so, with the searches attached, rather than filling the space with something that sounds like an answer.

An absence you can name is a research agenda. An absence nobody has named is just the way things are.

This article is educational and is not a substitute for care from someone who can look at your skin. Anything new, changing, painful or spreading should be assessed rather than researched.