Unequal Risk, Later Diagnosis: Dementia Across Ethnic and Racialised Groups

Dementia does not fall evenly across the population. In the US, the UK, Australia, Canada and the Netherlands, studies find higher rates of dementia among Black, Indigenous and several migrant and minority ethnic groups than among the majority population, and in the US and the UK people from these groups are less likely to be diagnosed, or are diagnosed later^[1][2][3].

For a family, those figures can mean memory changes put down to “old age” for longer, a memory test given in a second language, or a support group where no one shares the person’s language, faith or food. None of this comes from ethnicity itself. The research points to differences in lifelong health, education, income and housing, to racism and discrimination, and to services that were not designed with everyone in mind. This article sets out what the evidence shows, country by country, and where it is still missing.

In one large US health system, dementia rates differed by more than 60% between racial and ethnic groups

The clearest comparisons come from large US health systems where everyone is insured. Among 274,283 Kaiser Permanente health plan members in California aged 64 and over, followed from 2000 to 2013, new dementia diagnoses ran at 26.6 per 1,000 person-years for African-Americans and 22.2 for American Indian and Alaska Native members, against 19.3 for white and 15.2 for Asian-American members^[1]. The risk was 65% higher for African-Americans than Asian-Americans (hazard ratio 1.65), and rates varied by more than 60% between groups. The estimated 25-year cumulative risk of dementia from age 65 was 38% for African-Americans and 30% for white members^[1].

In a random sample of 1,869,090 veterans aged 55 and over cared for by the US Veterans Health Administration between 1999 and 2019, the age-adjusted rate of diagnosed dementia was 19.4 per 1,000 person-years for Black and 20.7 for Hispanic veterans, against 11.5 for White veterans. After adjustment, Black veterans had 1.54 times and Hispanic veterans 1.92 times the rate of diagnosis of White veterans^[4].

Direct testing tells the same story. In the nationally representative Health and Retirement Study’s 2016 cognitive assessment of people aged 65 and over, 10% had dementia overall, and dementia was more common among non-Hispanic Black people (odds ratio 1.81) than non-Hispanic White people; other groups were too small to compare^[5]. The Alzheimer’s Association summarises the US evidence as older Black Americans being about twice, and older Hispanic Americans about one and a half times, as likely as older White Americans to have Alzheimer’s or another dementia^[6].

Indigenous peoples in Australia and Canada face higher dementia rates, often at younger ages

Some of the largest gaps reported are among Indigenous peoples. A 2025 analysis of respondent-reported dementia among everyone aged 45 and over in Australia’s 2021 census found an age-standardised prevalence of 16.2 per 1,000 among Aboriginal and Torres Strait Islander people, against 7.4 per 1,000 among non-Indigenous people^[2]. The gap was widest in midlife: Aboriginal and Torres Strait Islander people aged 45 to 49 were 3.5 times as likely to report dementia as non-Indigenous people the same age, and those aged 50 to 54 3.8 times, narrowing to 1.6 times at 80 to 84. Odds of dementia fell with remoteness for non-Indigenous people, but not for Aboriginal and Torres Strait Islander people^[2].

In a study carried out in collaboration with urban and regional Aboriginal communities, 336 Aboriginal people aged 60 and over took part, with diagnoses made by a panel of clinicians; crude dementia prevalence was 13.4% and age-standardised prevalence 21.0%, most often Alzheimer’s disease or mixed dementia^[7]. Its authors note that prevalence among Aboriginal and Torres Strait Islander Australians has been found to be three to five times that of the general Australian population^[8].

In Canada, physician-treated dementia in Alberta in 2009 had an age-standardised prevalence of 7.5 per 1,000 among First Nations people, against 5.6 per 1,000 among other Albertans; it rose faster among First Nations people and affected younger age groups and men more^[9]. In Aotearoa New Zealand, there are still no national data on dementia prevalence, and national estimates rely on overseas data; a 2024 feasibility study for the first national prevalence study found that Pacific communities needed a different, culturally appropriate approach to recruitment and interviewing^[10].

In the UK and the Netherlands, Black and several migrant groups have higher rates

The UK has some of the deepest data on ethnicity and dementia in Europe, because large research databases of GP records include ethnicity, so we use it as a worked case. In UK primary care records for 2,511,681 people aged 50 to 105, followed from 2007 to 2015, the rate of new dementia diagnoses was 25% higher among Black women and 28% higher among Black men than among white women and men. It was 18% lower among Asian women and 12% lower among Asian men, though the difference for men was not statistically clear^[3]. The authors could not tell whether the lower Asian rate reflects less dementia or more missed diagnoses^[3].

A later study of English GP records linked to hospital and death data, covering people aged 65 and over from 1997 to 2018, found diagnosed dementia 1.22 times as common among Black people as among White people. South Asian and Black people with dementia were also diagnosed at younger ages, and died younger: mean age at death was about three years lower for South Asian people (2.97 years) and 2.66 years lower for Black people than for White people^[11]. In UK Biobank, a large volunteer cohort, Black participants had 1.43 times the risk of dementia of White participants, while South Asian participants had a similar risk^[12].

Evidence from mainland Europe is thinner, but the gaps can be large. In a Dutch study of 2,254 people aged 55 and over recruited through their GPs and screened with a test designed for people with little formal education, dementia was found in 14.8% of people of Turkish origin, 12.2% and 11.3% of Moroccan Arabic and Berber origin, and 12.6% of Surinamese-Hindustani origin, against 3.5% of native Dutch people. Surinamese-Creole participants (4.0%) were close to the native Dutch rate^[13].

The gaps track health, education, income and racism, not ethnicity itself

Ethnic and racial categories are social groupings, not biological ones. The studies point to an unequal spread of conditions associated with dementia risk. In English GP records for 865,674 people aged 65 and over, high blood pressure, abnormal blood fats (dyslipidaemia), obesity and diabetes were all more common in people from minority ethnic groups than in White people. The link between several of these factors and dementia was also stronger in South Asian people, and the link with high blood pressure stronger in Black people^[14]. In UK Biobank, by contrast, the main risk factors had similar associations with dementia in White, South Asian and Black participants^[12], so the evidence here is mixed.

Education runs through much of this. In the US national assessment, each extra year of schooling was associated with lower odds of dementia (odds ratio 0.93 per year)^[5], and in Australia having no educational attainment was strongly associated with dementia among both Indigenous and non-Indigenous people^[2]. Among older Aboriginal Australians, childhood trauma, unskilled work, stroke and head injury were independent predictors of dementia, which the researchers link to life-course social determinants of health^[8].

Racism itself is beginning to be measured. In Northern Manhattan, New York, a study of older residents found that Black adults were the group most likely to show negative effects of neighbourhood segregation on memory, language and dementia, which the authors link to the downstream effects of structural racism, such as unequal wealth and resources; the link with new dementia was weaker than with test scores^[15]. These are associations, not proof of cause. For the wider picture, see our studies on modifiable risk factors and deprivation and dementia.

Diagnosis comes later, and treatment less often

Higher diagnosis rates do not mean timely diagnosis. In the UK, combining GP diagnosis rates with community estimates of how many people develop dementia, researchers estimated that 42% of Black men developing dementia in 2015 had a diagnosis, against 53% of white men, so the higher diagnosis rate among Black people probably understates their higher risk^[3]. In the US, linking the Health and Retirement Study to Medicare records for people aged 70 and over, non-Hispanic Black people had roughly double the risk of having dementia but no diagnosis, compared with non-Hispanic White people^[16].

Timing matters too. Among 10,472 Medicare beneficiaries in California newly diagnosed with dementia or mild cognitive impairment in 2015, being diagnosed first at the earlier stage of mild cognitive impairment was treated as a timely diagnosis. Compared with White beneficiaries, Asian beneficiaries had 0.46 times, Hispanic beneficiaries 0.62 times and Black beneficiaries 0.73 times the odds of a timely diagnosis, even after allowing for neighbourhood disadvantage and rurality^[17].

After diagnosis, the gaps continue. In a long-running US research cohort (the National Alzheimer’s Coordinating Center data), where participants have access to dementia experts, Black and Hispanic participants with Alzheimer’s disease were still less likely than White participants to start dementia medicines, and less likely to stay on them; the authors suggest the gaps in the wider community are probably larger^[18]. A 2026 review of 17 studies found people with dementia from minority ethnic groups report lower-quality care and are less likely to be prescribed and receive anti-dementia medication^[19]. It names the barriers as including stigma, viewing dementia as a normal part of aging, limited language proficiency and services not tailored to minority ethnic groups^[19].

The tests themselves can be part of the problem. Standard screening tools such as the MMSE and MoCA often fail to identify cognitive impairment in people with low literacy or no formal education^[20], and NICE guidance for England recognises that a cognitive score may be inappropriate when a test cannot be given in a language in which the person is sufficiently fluent^[21]. For how diagnosis is counted, see our study on dementia diagnosis rates.

Many groups are barely counted, and almost absent from treatment trials

Even in high-income countries, whole groups are missing. US studies often cannot compare Asian, Pacific Islander, American Indian or Alaska Native people because samples are too small^[5]. A systematic review of 20 studies of health service use by minority ethnic people with dementia found 13 were from the US, four from the UK and one each from Australia, Belgium and the Netherlands, and that ethnicity and service use were recorded in very different ways^[22].

Treatment research is where the gap is starkest. Of 88 US phase 3 drug trials for Alzheimer’s disease run between 1997 and 2023, nearly half of the published trials (49.3%) did not report participants’ race or ethnicity at all^[23]. In the median trial, White patients made up 91.3% of those enrolled; non-Hispanic Black patients 7.2%, Hispanic patients 5.2%, Asian or Pacific Islander patients 0.9% and Native American patients 0.4%^[23]. Only 3 of 71 trials with results analysed outcomes by race or ethnicity, and the picture showed little improvement over time^[23].

Culturally adapted support is promising but small-scale. A 2026 rapid review of culturally adapted post-diagnostic support for South Asian people with dementia and their carers found twelve studies. Carer education programmes increased carers’ knowledge and confidence, and a South Asian link nurse improved access to services. But the research was small-scale and methodologically limited, with little focus on support for people living with dementia themselves^[24].

What the evidence asks of us

• Design services for the people most likely to be missed. NICE guidance for England asks service providers to make dementia services accessible to people less likely to use health and social care, naming people from Black, Asian and minority ethnic groups, and asks clinicians to secure equal access to treatment for people from different ethnic groups and cultural backgrounds^[21].
• Use assessments that work across languages and schooling. Tests that depend on reading, writing or fluency in the majority language miss impairment in people with little formal education^[20]. Interpreters and culturally adapted tools belong in standard memory services.
• Target prevention where risk is highest. High blood pressure, diabetes and obesity are more common in minority ethnic groups in England, and their associations with dementia appear stronger in some of them^[14]. Prevention should reach these communities first, shaped with them.
• Count everyone. Aotearoa New Zealand still has no national dementia prevalence data^[10], and many US and European studies are too small to describe Asian, Pacific and Indigenous groups^[5][22]. Health systems should record ethnicity in dementia data and publish diagnosis and treatment figures by ethnic group.
• Make trials representative. Alzheimer’s trials that enrol a median of 91.3% White patients, and rarely analyse results by group^[23], cannot show whether new treatments work equally well for everyone.
• Fund culturally adapted support, and test it properly. Link workers and adapted carer programmes show promise, but the evidence is small-scale^[24]. Indigenous-led and community-led services, like the community collaboration behind the Aboriginal study here^[7], need sustained funding and rigorous evaluation.

Where dementia is more common, diagnosis should be quicker, support closer to hand and research more inclusive. At present, the evidence shows the opposite.

Language: this piece uses person-first language ("people living with dementia", "people with dementia"), as most dementia organisations and people with lived experience prefer, and uses clinical terms such as "Alzheimer's disease" or "mild cognitive impairment" only where a study, dataset or diagnosis refers to that specific condition.

Why we're publishing this

We compiled this evidence because people living with dementia from Black, Asian, Indigenous, migrant and other minority ethnic communities, and the organisations that support them, are too often told that the data on their experience do not exist. Much of it does. Bringing it together gives charities, community groups and campaigners something to put in front of commissioners, health systems and research funders. The case is theirs; our part is to say plainly what the data show. This article is part of our Dementia & Health research library.

How to read this data

Most figures here are recorded diagnoses in health records (Kaiser Permanente, US veterans, UK and English GP records, Alberta). These depend on who is seen and diagnosed, so where a group is underdiagnosed its true rate is likely to be higher than recorded. The US Health and Retirement Study assessment, the Aboriginal community study and the Dutch study tested people directly; the Australian census figures are self- or family-reported. UK Biobank volunteers are not representative of the population. Rates are expressed in different ways (per 1,000 person-years, per 1,000 people, percentages, hazard, incidence rate and odds ratios) and cover different ages, years and countries, so they are not directly comparable with each other. Ethnic categories also differ between countries and studies: a broad label such as “Asian” or “Hispanic” groups together very different populations, and the same label can mean different things in different countries. Studies show associations; they do not show that ethnicity causes dementia. The evidence comes overwhelmingly from high-income countries.

Use this data

Free to cite with attribution to Understanding Aging. Charities, carers' groups, 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 and statistics for general information. It is not medical, care or financial 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 memory or health, please speak to a GP or another qualified health professional.

Sources

1. Mayeda ER, Glymour MM, Quesenberry CP, Whitmer RA. Inequalities in dementia incidence between six racial and ethnic groups over 14 years. Alzheimer’s & Dementia, 2016. europepmc.org/article/MED/26874595
2. Clarke AJ, Ahmed M, Katzenellenbogen JM, et al. The intersection of rurality and dementia prevalence in Australia for Aboriginal and Torres Strait Islander and non-Indigenous peoples. Medical Journal of Australia, 2025. europepmc.org/article/MED/40313188
3. Pham TM, Petersen I, Walters K, et al. Trends in dementia diagnosis rates in UK ethnic groups: analysis of UK primary care data. Clinical Epidemiology, 2018. europepmc.org/article/MED/30123007
4. Kornblith E, Bahorik A, Boscardin WJ, et al. Association of race and ethnicity with incidence of dementia among older adults. JAMA, 2022. europepmc.org/article/MED/35438728
5. Manly JJ, Jones RN, Langa KM, et al. Estimating the prevalence of dementia and mild cognitive impairment in the US: the 2016 Health and Retirement Study Harmonized Cognitive Assessment Protocol Project. JAMA Neurology, 2022. europepmc.org/article/MED/36279130
6. Alzheimer’s Association. Alzheimer’s Disease Facts and Figures (2026 edition, summary). Alzheimer’s Association, 2026. alz.org/alzheimers-dementia/facts-figures
7. Radford K, Mack HA, Draper B, et al. Prevalence of dementia in urban and regional Aboriginal Australians. Alzheimer’s & Dementia, 2015. europepmc.org/article/MED/24985534
8. Radford K, Lavrencic LM, Delbaere K, et al. Factors associated with the high prevalence of dementia in older Aboriginal Australians. Journal of Alzheimer’s Disease, 2019. europepmc.org/article/MED/30507573
9. Jacklin KM, Walker JD, Shawande M. The emergence of dementia as a health concern among First Nations populations in Alberta, Canada. Canadian Journal of Public Health, 2012. europepmc.org/article/MED/23618107
10. Martinez-Ruiz A, Yates S, Cheung G, et al. Living with Dementia in Aotearoa (LiDiA): a feasibility study for a dementia prevalence study in Māori and non-Māori living in New Zealand. Dementia, 2024. europepmc.org/article/MED/37137731
11. Mukadam N, Marston L, Lewis G, et al. Incidence, age at diagnosis and survival with dementia across ethnic groups in England: a longitudinal study using electronic health records. Alzheimer’s & Dementia, 2023. europepmc.org/article/MED/36047605
12. Mukadam N, Marston L, Lewis G, Livingston G. Risk factors, ethnicity and dementia: a UK Biobank prospective cohort study of White, South Asian and Black participants. PLOS ONE, 2022. europepmc.org/article/MED/36223334
13. Parlevliet JL, Uysal-Bozkir Ö, Goudsmit M, et al. Prevalence of mild cognitive impairment and dementia in older non-western immigrants in the Netherlands: a cross-sectional study. International Journal of Geriatric Psychiatry, 2016. europepmc.org/article/MED/26799690
14. Mukadam N, Marston L, Lewis G, et al. South Asian, Black and White ethnicity and the effect of potentially modifiable risk factors for dementia: a study in English electronic health records. PLOS ONE, 2023. europepmc.org/article/MED/37819899
15. Pohl DJ, Seblova D, Avila JF, et al. Relationship between residential segregation, later-life cognition, and incident dementia across race/ethnicity. International Journal of Environmental Research and Public Health, 2021. europepmc.org/article/MED/34769752
16. Gianattasio KZ, Prather C, Glymour MM, et al. Racial disparities and temporal trends in dementia misdiagnosis risk in the United States. Alzheimer’s & Dementia: Translational Research & Clinical Interventions, 2019. europepmc.org/article/MED/31890853
17. Tsoy E, Kiekhofer RE, Guterman EL, et al. Assessment of racial/ethnic disparities in timeliness and comprehensiveness of dementia diagnosis in California. JAMA Neurology, 2021. europepmc.org/article/MED/33779684
18. Zhu CW, Neugroschl J, Barnes LL, Sano M. Racial/ethnic disparities in initiation and persistent use of anti-dementia medications. Alzheimer’s & Dementia, 2022. europepmc.org/article/MED/35218291
19. Aguzzoli E, Walbaum M, Knapp M. Access to healthcare services and adherence to treatments for people with dementia among ethnic minority groups: a scoping review. Frontiers in Dementia, 2026. europepmc.org/article/MED/41777856
20. Jilla M, Mars S, Laghnimi N, et al. Culturally adapted cognitive screening: validation of the Jilla Moroccan Dementia Scale. Dementia and Geriatric Cognitive Disorders, 2026. europepmc.org/article/MED/41231739
21. National Institute for Health and Care Excellence. Dementia: assessment, management and support for people living with dementia and their carers (NG97), Recommendations. NICE, 2018. nice.org.uk/guidance/ng97/chapter/Recommendations
22. Co M, Couch E, Gao Q, et al. Access to health services in older minority ethnic groups with dementia: a systematic review. Journal of the American Geriatrics Society, 2021. europepmc.org/article/MED/33230815
23. Lin Z, Sun R, Ross JS, et al. Racial and ethnic reporting and representation in US Alzheimer clinical trials: a systematic review. JAMA Network Open, 2026. europepmc.org/article/MED/41893846
24. Akhtar A, Nwofe E, Dodd E, et al. Culturally adapted post-diagnostic dementia support for South Asian people living with dementia and caregivers: a rapid review. Age and Ageing, 2026. europepmc.org/article/MED/42720437