more_vert
Despite the high prevalence of melanoma in Australia, a new article in The Medical Journal of Australia argues people of colour are under-represented in the registry data used to collect information on skin cancer.
The lack of Australian data on skin cancer in people of colour can have downstream impacts on a clinician’s ability to diagnose the disease early and accurately, according to a Perspective published in The Medical Journal of Australia today.
Melbourne dermatologist Dr Michelle Rodrigues and Monash University PhD candidate Mr Ayooluwatomiwa Oloruntoba write that while Australia has one of the world’s highest incidence and mortality rates for melanoma (here and here), it is also one of the most ethnically diverse countries in the world.
“When I realised our skin cancer registries did not capture information on a patients’ skin colour and ethnicity, I felt the need to shed some light on the issue,” Dr Rodrigues told InSight+.
“The skin cancer registries were set up at a time when cutaneous diversity was not widespread in Australia.”
Dr Rodrigues and Mr Oloruntoba wrote that it is currently unclear to what extent Australians with skin of colour are affected by melanoma and keratinocyte skin cancer.
They argue recording skin phototype, immigration status and country of birth, alongside racial and ethnic background (information they have termed “SIRE data”) in skin cancer registries will be advantageous in many ways.
“By capturing ethnicity and skin colour, I hope we will be able to identify how to better care for Australians with skin of colour who are diagnosed with skin cancer,” Dr Rodrigues said.
“Registries in the United States show that skin cancer in patients with skin of colour often presents at more advanced stages of disease and this is associated with an increased risk in mortality compared with patients of European ancestry [here and here],” they wrote.
In an interview with InSight+, Professor Anne Cust, Chair of Cancer Council’s Skin Cancer Committee, agreed that data collection of Australians with skin of colour would be helpful to doctors.
“As a researcher and epidemiologist, I believe collecting this data would be really useful, so clinicians get a better understanding of what is happening with a patient’s skin and can improve their care,” Professor Cust said.
“While [people of colour] may have fewer skin cancers develop, when they do develop a skin cancer their outcomes appear to be worse.”
Professor Cust also said the development of artificial intelligence (AI) software that can detect skin cancers must consider the needs of people of colour.
“There is a lot of research around using software that automatically detects when a legion is a melanoma or a different type of skin cancer,” Professor Cust said.
“The way the algorithms work is that they are trained on datasets, [but] it is all depends on what datasets these algorithms are trained on.
“If [the only dataset is] white skin lesions developing on white skin, then obviously the algorithms are not going to be as accurate in detecting skin cancer on someone with skin of colour.
“The use of AI … is not very widespread at the moment, it is very much still being researched.”
Dr Rodrigues said that the use of AI has the potential to improve the detection of skin cancers in all Australians.
“I see AI as a tool to triage lesions of concern, to assist the clinician in a clinical course of action and believe its utility is not yet completely appreciated,” she said.
“AI will help us make faster and more accurate diagnoses, which can then be paired with the clinical judgment and experience of the dermatologists or practitioner to help our patients navigate next steps in their treatment journey.”
Professor Cust explained that skin cancer data were collated by the Australian Institute of Health and Welfare, which maintains the Australian Cancer Database.
“The skin cancer data collected by the states and territories are all different, but they will all contain a core dataset,” Professor Cust told InSight+.
“So [the states and territories] collect data on melanomas, but they don’t collect all data on non-melanomas. We know very little about non-melanoma skin cancers. The data collected for melanoma does not include skin colour or skin type, but it will include country of birth and Indigenous status.”
Dr Rodrigues proposed that SIRE data be ethically obtained and recorded in registries going forward.
Professor Cust agreed, saying the Cancer Council was continuing its work on researching the benefits and harms of sun exposure.
Subscribe to the free InSight+ weekly newsletter here. It is available to all readers, not just registered medical practitioners.