Ethnicity and Cancer: Unequal Diagnosis, Care and Survival

Across countries with good data, people from some minority ethnic groups and Indigenous peoples are more likely to be diagnosed late, to report poorer care and to die of their cancer than the majority population, often even where the cancer itself is no more common. The sources that measure this point mainly to differences in access, risk-factor exposure and the social conditions people live in, not to biology.

For most people, this does not show up as a single dramatic moment. It is the extra wait before a symptom is taken seriously, the appointment where nobody checks that information has been understood, the form where ethnicity is left blank so the gap never appears in the figures. Each of these is a feature of a service, not of the person using it, and each can be measured and changed.

In England, most cancers are less common in minority ethnic groups, with some clear exceptions

The most complete national picture of incidence by ethnicity comes from England, which records ethnicity in its hospital and cancer datasets and can therefore be used as a worked case. A national analysis of almost 3 million diagnoses across 31 cancer sites in 2013–2017 compared age-standardised incidence rates for the broad Asian, Black, Mixed/Multiple and White ethnic groups^[2]. For most cancer sites, and for most combinations of ethnic group and sex, rates were lower in non-White minority ethnic groups than in the White group. Rate ratios were particularly low, below 0.5, for melanoma skin cancer and for some cancers strongly linked to smoking, including lung, bladder and oesophageal cancer^[2].

The exceptions matter. Prostate cancer incidence was 2.1 times higher in Black males than White males, and myeloma 2.7–3.0 times higher in people of Black ethnicity. Several gastrointestinal cancers were 1.1–1.9 times higher in the Black group and 1.4–2.2 times higher in the Asian group, and thyroid cancer and Hodgkin lymphoma were also more common in some groups^[2]. Because of prostate cancer, the all-cancers rate for Black males was 14% higher than for White males. Prostate cancer risk in Black men is covered in depth in our article on prostate cancer.

The authors are explicit about what they think drives these differences: “Exogenous risk factors are likely to explain the bulk of the difference in cancer incidence rates between ethnic groups”, with differences in risk-factor prevalence and in access to and use of health services playing a larger part than genetic factors^[2]. Lower incidence is not the same as better outcomes, either: as the sections below show, a group can be less likely to develop a cancer and still more likely to be diagnosed late or to die from it.

Some groups are more often diagnosed at a late stage, and the route in does not fully explain it

Stage at diagnosis is one of the strongest influences on survival, and it is where gaps in access show up first. A national analysis of English registry and hospital records from 2012–2016 compared White British, Caribbean, African, Chinese and Asian people aged 15–99 across six cancer sites, adjusting for confounding factors^[4]. Compared with the White British group, Caribbean, African and Asian women with breast or ovarian cancer, and Caribbean and African women with uterine or colon cancer, had higher odds of being diagnosed at stage 3 or 4. So did Caribbean women with non-small cell lung cancer and Caribbean men with colon cancer. The pattern was not uniform: Caribbean and African men with prostate cancer had lower odds of a late-stage diagnosis^[4]. The authors point to low symptom awareness and barriers to presentation and referral as causes of delay, and to removing referral barriers as part of the answer^[4].

How people reach a diagnosis tells a more mixed story. A UK cohort of 243,825 people with ten cancers diagnosed in 2006–2016 found that, across all groups, 36.4% were diagnosed through the urgent two-week-wait referral route and 18.2% through an emergency presentation^[5]. The “Other” ethnic group had the highest share of emergency diagnoses, followed by White patients, while Asian and Black patients were more likely to be referred by their GP and Black and Mixed patients more likely to follow the two-week-wait route. The authors concluded that where Asian and Black patients do have worse cancer outcomes, these “are unlikely to be arising solely from a poorer diagnostic process”^[5]. In other words, the problem may lie as much in what happens before people present, and after diagnosis, as in the referral itself. The wider evidence on stage is in our article on stage at diagnosis.

In the US, Black and Native American people die of cancer at higher rates than incidence alone would predict

The United States publishes cancer statistics using federal race and ethnicity categories, with race reported exclusive of Hispanic ethnicity. The American Cancer Society’s 2025 report on African American and Black people found that, over the most recent five years of data, Black men had 16% higher cancer mortality than White men despite only 4% higher incidence, and Black women had 10% higher mortality than White women despite 9% lower incidence^[1]. Death rates among Black people were about twice those of White people for prostate, uterine corpus and stomach cancers and for myeloma, and 40–50% higher for colorectal, breast, cervical and liver cancers. Black people have the lowest survival of any racial or ethnic group for most cancers^[1].

Breast cancer shows the gap most starkly. Non-Hispanic Black women are slightly less likely than non-Hispanic White women to be diagnosed with breast cancer (a rate ratio of 0.95), but their breast cancer death rate is 38% higher: 26.8 against 19.4 deaths per 100,000 women^[1]. The report’s authors attribute these disparities largely to “inequalities in the social determinants of health that trace back to structural racism”^[1].

Native American people have the highest cancer mortality of any group in the US, with death rates for kidney, liver, stomach and cervical cancers two to three times those of White people, according to the 2025 national cancer statistics report (mortality data through 2022)^[6].

Coverage status has been examined as one factor. A study of 54,558 children and adolescents diagnosed with cancer at age 19 or under in 2004–2010, using the National Cancer Database, found that racial and ethnic minority groups had a 14–42% higher hazard of death than non-Hispanic White children, with the size of the gap varying by insurance type^[7]. For non-Hispanic Black children the hazard ratio was 1.48 among those privately insured and 1.68 among those uninsured. The authors conclude that survival disparities exist across insurance types and point to the need for efforts on health equity alongside improving coverage^[7].

Indigenous peoples in New Zealand and Australia face the same pattern of poorer survival

New Zealand records ethnicity on its national cancer registry, which allows outcomes for Māori to be compared with those of European New Zealanders. Among people diagnosed with blood cancers in 2007–2019 (2,653 Māori and 20,458 European people), Māori had poorer cancer-specific survival across most blood cancer types^[3]. After adjustment for age and sex, the hazard of dying from leukaemia was 1.77 times higher, from non-Hodgkin lymphoma 1.71 times higher and from myeloma 1.40 times higher. For Hodgkin lymphoma the estimate (1.18) was too uncertain to show a clear difference^[3]. Those ratios are adjusted for age and sex only, so they describe the size of the gap, not what lies behind it.

For lung cancer, a national project identified higher emergency presentation rates, poorer access to early detection and lower surgery rates among Māori, and concluded that access to timely, best-practice diagnosis and care “could be an important driver of these disparities”^[8]. In the Australian state of Victoria, a registry study of people diagnosed with lung cancer in 2008–2022 found that Aboriginal people (512 people, compared with 43,468 non-Aboriginal people) were younger, more likely to live outside a major city and had a modestly higher hazard of death from any cause (hazard ratio 1.12, 95% confidence interval 1.01–1.24)^[9]. Its authors describe the inequity as “an ongoing manifestation of colonisation”^[9]. These are single-country, single-cancer studies and their figures are not directly comparable with each other or with the US and English data above, but the direction is consistent.

People from minority ethnic groups report poorer experiences of cancer care

Experience of care is measured differently from survival, but it shapes whether people come back, ask questions and stay on treatment. England’s National Cancer Patient Experience Survey links responses to ethnicity. A secondary analysis of 25,408 women with breast cancer in the 2017/2018 survey found that, compared with White British women, Asian women (odds ratio 0.52), Black women (0.67) and women in the White Other group (0.63) had lower odds of rating their overall care positively^[10]. Similar patterns ran across every stage of the care pathway. The authors identified prolonged help-seeking, poor communication between patients and professionals, and variable access to professionals and resources as contributing factors^[10].

The finding is not confined to England. A systematic review of 36 studies covering 292,345 people with cancer, searched to August 2023, found that 23 studies showed worse experiences among racial and ethnic minority groups than among non-Hispanic White patients, and nine found worse experiences among people who preferred a language other than English^[11]. The reviewers note that definitions and measures of patient experience vary widely, but most of the studies that compared groups pointed the same way.

Gaps in ethnicity recording hide part of the inequality

Every figure in this article depends on ethnicity being recorded, recorded accurately and recorded in categories that mean something. That is not guaranteed anywhere.

• Missing records. In the English incidence analysis, 5.6% of diagnoses had ethnicity “Not Known” and were left out of the main rate calculations because there was no matching population count^[2]. A sensitivity analysis that shared the missing cases out across the recorded groups, in proportion to the known cases, changed rate ratios by at most 2 percentage points; a separate test of possible errors in the population estimates used as denominators moved them by up to 12 percentage points in the Black group^[2].
• Groups too broad to act on. The same study did not calculate rates for the “Other” ethnic group, about 1% of cases, because it was judged too mixed for results to translate into policy^[2]. Yet the UK routes-to-diagnosis study found that this same group had the highest share of emergency diagnoses^[5]. Broad categories such as “Asian” or “Black” can also hide large differences between communities, which is why the stage study used finer groups such as Caribbean and African^[4].
• Misclassification. US national statistics report race exclusive of Hispanic ethnicity “to minimize racial misclassification”, and adjust American Indian and Alaska Native mortality figures for racial misclassification^[12]. Those corrections exist because, left alone, misclassification would distort the very figures used to measure inequality.

Categories also differ between countries: England uses census-based ethnic groups, the US uses federal race and Hispanic-origin categories, and New Zealand and Australia record Indigenous status. Figures from one system cannot be mapped onto another, and this article does not try to. Where ethnicity is not recorded at all, inequality does not disappear; it simply goes unmeasured. Ethnic inequality also overlaps with the gaps by area deprivation described in our article on deprivation and cancer.

What the evidence asks of us

The evidence above locates the problem in the path to and through care, not in the people taking it. That points to practical action.

• Record ethnicity completely and consistently, using categories fine enough to act on, so that no group falls into “Not Known” or “Other” and out of view^[2].
• Treat ethnicity as an equity target, not only a variable. NHS England’s Core20PLUS5 approach names ethnic minority communities among the groups local services are expected to identify, and sets an early-diagnosis ambition of 75% of cancers diagnosed at stage 1 or 2 by 2028^[13]. Meeting it will require progress in every group, reported group by group.
• Remove barriers before and after referral. The stage study calls for work on symptom awareness and on barriers to presentation and referral^[4]; the New Zealand lung cancer work points to early detection and surgery as the parts of the pathway where Māori fare worse^[8].
• Measure and act on experience. Communication between patients and professionals, and language, are where minority groups report the widest gaps^[10][11]; services should publish experience results by ethnicity and show what they change in response.
• Address the social conditions behind the figures, which the US evidence traces to structural racism^[1], as part of cancer planning rather than separately from it.

Charities and community groups are often closest to the people these figures describe. The data give them grounds to ask, of every service they work with, whether outcomes are reported by ethnicity and what is being done where the gaps are widest. More on how cancer outcomes are measured is in the Cancer research library.

Language: this piece says "people with cancer" and "people living with and after cancer", and describes the illness and its treatment plainly, without combat metaphors. Clinical terms such as "stage", "incidence" and "net survival" are used in the sense the cited source defines them.

Why we're publishing this

Charities, community organisations and people affected by cancer from minority ethnic and Indigenous communities have long described these gaps from experience. The registry data and surveys bear them out. We have gathered the strongest figures we could verify, from several countries, so that the case for fairer cancer services can be made with evidence that is easy to cite and hard to dismiss.

How to read this data

Each figure uses its source’s own ethnicity categories, and we have not merged or relabelled them: England’s broad and detailed ethnic groups, US non-Hispanic race categories and Hispanic origin, and Indigenous status in New Zealand and Australia. Incidence and mortality comparisons are age-standardised rate ratios; survival comparisons are hazard ratios adjusted for age and sex (New Zealand) or stratified by age, sex and cancer subtype (Victoria), not for stage or deprivation; experience findings are odds ratios, which are not the same as risks. The stage and routes-to-diagnosis studies cover people diagnosed up to 2016, and the experience analysis uses the 2017/18 survey round; more recent rounds of that survey have been published since, and some patterns may have changed. Most of the evidence comes from high-income countries with good ethnicity recording; for much of the world there are no ethnicity-specific cancer data at all. Differences between groups describe populations, not individuals, and the sources cited attribute them mainly to risk-factor exposure, access to care and social conditions rather than to biology.

Use this data

Free to cite with attribution to Better Health For Everyone. Charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research for general information. It is not medical advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's health, please speak to a qualified health professional.

Sources

1. Saka AH, Giaquinto AN, McCullough LE, Tossas KY, Star J, Jemal A, Siegel RL. Cancer statistics for African American and Black people, 2025. CA: A Cancer Journal for Clinicians, 2025. europepmc.org/article/PMC/PMC11929131
2. Delon C, Brown KF, Payne NWS, Kotrotsios Y, Vernon S, Shelton J. Differences in cancer incidence by broad ethnic group in England, 2013–2017. British Journal of Cancer, 2022. europepmc.org/article/PMC/PMC9174248
3. Clough S, Wheeler M, Stanley J, Signal V, Ruka M, Koea J, Gurney J. Blood cancer incidence, mortality and survival for Māori in New Zealand. Cancer Epidemiology, 2024. europepmc.org/article/MED/39217827
4. Fry A, White B, Nagarwalla D, Shelton J, Jack RH. Relationship between ethnicity and stage at diagnosis in England: a national analysis of six cancer sites. BMJ Open, 2023. europepmc.org/article/PMC/PMC9884890
5. Martins T, Abel G, Ukoumunne OC, Mounce LTA, Price S, Lyratzopoulos G, Chinegwundoh F, Hamilton W. Ethnic inequalities in routes to diagnosis of cancer: a population-based UK cohort study. British Journal of Cancer, 2022. europepmc.org/article/PMC/PMC9427836
6. Siegel RL, Kratzer TB, Giaquinto AN, Sung H, Jemal A. Cancer statistics, 2025. CA: A Cancer Journal for Clinicians, 2025. europepmc.org/article/PMC/PMC11745215
7. Wang X, Brown DS, Cao Y, Ekenga CC, Guo S, Johnson KJ. Health insurance coverage and racial/ethnic disparities in US childhood and adolescent cancer survival. The Journal of Pediatrics, 2023. europepmc.org/article/MED/36889628
8. Gurney J, Stanley J, Davies A, Signal V, Dawkins P, Cameron L, et al. Improving lung cancer survival outcomes for Māori. New Zealand Medical Journal, 2025. europepmc.org/article/MED/40638933
9. Bergin ART, Te Marvelde L, Milne RL, et al. Sovereignty and survival: understanding lung cancer outcomes for Aboriginal and Torres Strait Islander people in Australia, a case-control study. Medical Journal of Australia, 2026. europepmc.org/article/MED/42764147
10. Estupiñán Fdez de Mesa M, Marcu A, Ream E, Whitaker KL. Using the Candidacy Framework to understand individual, interpersonal, and system level factors driving inequities in women with breast cancer: a cross-sectional study. BJC Reports, 2024. europepmc.org/article/PMC/PMC11524000
11. Koo CRJ, Mendoza M, Shah K, McGonigle R, Sharaf S, Matsoukas K, et al. Differences in cancer patient experience by race, ethnicity, and preferred language: a systematic review. Cancer Medicine, 2026. europepmc.org/article/MED/41999109
12. Siegel RL, Kratzer TB, Wagle NS, Sung H, Jemal A. Cancer statistics, 2026. CA: A Cancer Journal for Clinicians, 2026. europepmc.org/article/PMC/PMC12798275
13. NHS England. Core20PLUS5 (adults): an approach to reducing healthcare inequalities. NHS England, accessed 2026. england.nhs.uk/about/equality/equality-hub/…/core20plus5