Research and analysis

Ethnic disparities in the major causes of mortality and their risk factors – a rapid review

Updated 28 April 2021

Ethnic disparities in the major causes of mortality and their risk factors in the UK – submission to the Commission on Race and Ethnic Disparities.

Raghib Ali, Avirup Chowdhury, Nita Forouhi, Nick Wareham. MRC Epidemiology Unit, University of Cambridge.

This paper mainly focuses on the 2 leading causes of death in the UK: cancers and cardiovascular diseases which account for 55% of deaths in the UK, and their major risk factors.

In considering disparities, we look at both ethnicity and deprivation as there are strong associations between ethnicity and deprivation, and between deprivation and most health outcomes.

Social disparity by ethnic group

Figure 1: Distribution of population for each ethnic group by deprivation decile, England 2011 ^{[footnote 1]}

UK statistics show that ^{[footnote 2]}:

• people in Bangladeshi, Pakistani and Black ethnic groups are the most likely to be living in deprived neighbourhoods
• unemployment rates are highest among Black, Bangladeshi, and Pakistani populations, while White and Indian groups are more likely to be in employment
• people in Bangladeshi, Pakistani, Chinese and Black ethnic groups are about twice as likely to be living on a low income, and experiencing child poverty, as White people – most groups had a higher proportion of women in low pay than men, with a stark gender difference for White people (31% of women earning below the living wage compared with 16% for men)
• people from ethnic minorities are more likely to live in private rented accommodation than White British people (one-third vs 18%), and in overcrowded households (13.5% vs 2.8%), with 30.2% of households in the Bangladeshi group being overcrowded

Summary tables of disparities

Table 1. Life expectancy and mortality compared with the White ethnic group (England and Scotland)

Outcome or disease	South Asian ethnic group	Indian ethnic group	Pakistani ethnic group	Bangladeshi ethnic group	Black ethnic group	Black African ethnic group	Black Caribbean ethnic group	Chinese	Social gradient (Increased mortality with deprivation)
Life expectancy [footnote 3] (Men)	N/A	Significantly better (S)	Significantly better (S)	No significant difference (S)	N/A	No significant difference (S)	N/A	Significantly better (S)	Y
Life expectancy (Women)[footnote 3]	N/A	Significantly better (S)	Significantly better (S)	No significant difference (S)	N/A	No significant difference (S)	N/A	Significantly better (S)	Y
Healthy life expectancy (Men) [footnote 4]	N/A	Significantly better (S)	No significant difference (S)	N/A	N/A	No significant difference (S)	N/A	Significantly better (S)	Y
Healthy life expectancy (Women) [footnote 4]	N/A	No significant difference (S)	Significantly worse (S)	N/A	N/A	No significant difference (S)	N/A	Significantly better (S)	Y
Overall mortality (Men)#[footnote 5]	Significantly better	N/A	N/A	N/A	Significantly better	N/A	N/A	N/A	Y
Overall mortality (Women)# [footnote 5]	Significantly better	N/A	N/A	N/A	Significantly better	N/A	N/A	N/A	Y
Premature mortality (age under 75) (Men)*[footnote 1]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	No significant difference	Significantly better	Y
Premature mortality (age under 75) (Women)[footnote 1]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	No significant difference	Significantly better	Significantly better	Y

Table notes: (S): based on Scotland data only^{[footnote 3]}, N/A: data not available, #Based on ethnicity in England ^{[footnote 5]}, *Based on country of birth (for England) ^{[footnote 1]}

Table 2. The 25 leading causes of mortality as measured by years of life lost, compared with the White ethnic group (England, 2016) ^{[footnote 6]}

Causes of mortality in England ranked by years of life lost (YLL), age-standardised	South Asian ethnic group	Indian ethnic group	Pakistani ethnic group	Bangladeshi ethnic group	Black ethnic group	Black African ethnic group	Black Caribbean ethnic group	Chinese	Increased mortality with deprivation
All cancers[footnote 1] [footnote 7]#	Significantly better	Significantly better	Significantly better	Significantly better	No significant difference	No significant difference	No significant difference	Significantly better	Y
Ischaemic heart disease[footnote 1] [footnote 8] [footnote 10]	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly better	Significantly better	No significant difference	Significantly better	Y
Lung cancer[footnote 7]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Stroke[footnote 8] [footnote 11] [footnote 12]	Significantly worse	No significant difference	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly better	Y
Chronic Obstructive Pulmonary Disease[footnote 13] [footnote 14]	Significantly better	Significantly better [S]	No significant difference [S]	No significant difference [S]	Significantly better	No significant difference [S]	N/A	Significantly better	Y
Dementia/ Alzheimer’s disease[footnote 15]	Significantly better	N/A	N/A	N/A	Significantly worse	N/A	N/A	N/A	Y
Lower Respiratory Tract Infection[footnote 16]*	N/A	No significant difference [S]	Significantly worse [S]	Significantly worse [S]	N/A	No significant difference [S]	N/A	Significantly better [S]	Y
Self-harm or suicide[footnote 1] [footnote 17]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Colorectal cancer[footnote 7] [footnote 18]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	No significant difference	Y
Breast cancer[footnote 7] [footnote 19]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Neonatal preterm birth[footnote 1]	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	No significant difference	Y
Congenital defects[footnote 20]	Significantly worse	No significant difference	Significantly worse	No significant difference	Significantly worse	No significant difference	No significant difference	N/A	Y
Other cardiovascular disease [footnote 21]	N/A	N/A	Significantly worse [S]	N/A	N/A	N/A	N/A	Significantly better [S]
Pancreatic cancer[footnote 7] [footnote 18]	Significantly better	Significantly better	Significantly better	Significantly better	No significant difference	No significant difference	No significant difference	No significant difference	Y
Road injuries[footnote 22] [footnote 23]	Significantly better	No significant difference [S]	No significant difference [S]	N/A	Significantly worse	No significant difference [S]	N/A	No significant difference [S]	Y
Drug use disorders[footnote 24]	Significantly better	N/A	N/A	N/A	No significant difference	N/A	N/A	N/A	Y
Other cancers[footnote 7]	Significantly better	Significantly better	Significantly better	Significantly better	No significant difference	No significant difference	No significant difference	Significantly better	Y
Cirrhosis alcohol[footnote 25] [footnote 26]	N/A	Significantly worse	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Oesophageal cancer[footnote 7] [footnote 18]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Brain Cancer [footnote 7] [footnote 27]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	N
Prostate cancer[footnote 7] [footnote 28]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly worse	Significantly worse	Significantly worse	Significantly better	N
Leukaemia[footnote 7] [footnote 29]	Significantly better	Significantly better	No significant difference	Significantly better	No significant difference	No significant difference	No significant difference	No significant difference	N
Stomach cancer[footnote 7] [footnote 18]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly worse	No significant difference	Significantly worse	Significantly worse	Y
Lymphoma[footnote 7] [footnote 29]	Significantly better	Significantly better	No significant difference	Significantly better	No significant difference	Significantly worse	Significantly better	No significant difference	Y
Ovarian cancer[footnote 7] [footnote 19]	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	N
Cirrhosis Hep C [footnote 25] [footnote 26]	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Y

Table notes: (S): based on Scotland data only, N/A: data not available, #:All cancers combined is not included in the Public Health England analysis but it is the leading cause of death in the UK; the data does not include COVID-19 because it is for time periods before 2020

Table 3. 12 main risk factors for combined mortality and morbidity in disability adjusted life years, compared with the White ethnic group (England, 2016) ^{[footnote 6] [footnote 30]}

Main risk factors for DALYs in England, age-standardised	South Asian ethnic group	Indian ethnic group	Pakistani ethnic group	Bangladeshi ethnic group	Black ethnic group	Black African ethnic group	Black Caribbean ethnic group	Chinese	Increased burden with deprivation
Tobacco	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Dietary risks	N/A	N/A	N/A	N/A	N/A	N/A	N/A	No significant difference	N/A
High BMI (adults)	Significantly better	N/A	N/A	N/A	Significantly worse	N/A	N/A	Significantly better	Y
High BMI (children)	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly better	Y
High blood pressure	Significantly better	Significantly better	Significantly better	Significantly better	Significantly worse	Significantly worse	Significantly worse	Significantly better	Y
Alcohol	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
Drug use	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Significantly better	Y
High total cholesterol	Significantly worse	N/A	N/A	N/A	Significantly better	N/A	N/A	N/A	Y
Occupational risks	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	Y
High plasma glucose	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Y
Air pollution [footnote 31]	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse	Significantly worse		Y
Low physical activity	Significantly worse	N/A	N/A	N/A	Significantly worse	N/A	N/A	Significantly worse	Y

Table notes – N/A: data not available

Outcomes, risk factors and ‘causes of the causes’

Table 4 below illustrates the relationships across the life course between risk factors and outcomes for the 5 main causes of death and disability in the UK.

It shows, in a simplified form, how proximal and intermediate risk factors (which are often behavioural and in theory modifiable by individuals) are in turn affected by distal risk factors or the ‘causes of the causes’.

Some of these are fixed (for example, genetics) whereas others can be changed (for example, socio-economic factors) but often require government or societal action to facilitate that change.

Although interventions can be implemented at all ‘distances’ of risk factors, the earlier they are applied in the life course, the greater the impact on prevention of the outcome.

Table 4: The relationship between the causes and outcomes of the five main causes of death and disability

Distal risk factors (‘Causes of the causes’)	Intermediate risk factors (modifiable or behavioural)	Proximal risk factors (also outcomes)	Outcomes (5 main causes of death and disability)
Socio-economic factors Genetics Epigenetics Exposures in the womb and early life	Tobacco Alcohol Unhealthy diet Physical inactivity Air pollution	High blood pressure High cholesterol High glucose Obesity	Heart disease Cancer Chronic lung diseases Diabetes Mental health disorders

Life expectancy, healthy life expectancy, overall mortality and premature mortality

Life expectancy is an important summary measure of mortality, providing information on overall trends in population health, and reflects the impact of the key determinants of health (socio-economic, education, income, housing, employment) over the whole life course and so is a key measure of overall health.

Figure 2: Life expectancy at birth by deprivation in England ^{[footnote 1]}

Table 5: Highest and lowest life expectancies by local areas in the UK

Area (women)	Country (women)	Life expectancy (women)	Area (men)	Country (men)	Life expectancy (men)
Highest			Highest
Westminster	England	87.22	Westminster	England	84.88
Camden	England	87.10	Kensington and Chelsea	England	83.86
Kensington and Chelsea	England	86.96	Camden	England	83.27
Epsom and Ewell	England	86.88	Harrow	England	83.20
Mole Valley	England	86.44	Hart	England	83.13
Lowest			Lowest
Blackpool	England	79.54	West Dunbartonshire	Scotland	75.06
Dundee City	Scotland	79.48	Inverclyde	Scotland	74.87
West Dunbartonshire	Scotland	79.21	Blackpool	England	74.40
Inverclyde	Scotland	79.01	Dundee City	Scotland	73.92
Glasgow City	Scotland	78.50	Glasgow City	Scotland	73.60

The lowest life expectancies among local areas are found in Scotland, while the highest are found in London ^{[footnote 32]}

These areas are mainly populated by White people, indicating that there are likely to be significant differences within the ‘White’ ethnic group – that is, between White people who live in the 10% most deprived areas of England (predominantly in the North and coastal towns) and those in the 10% least deprived (mainly in London and the South East). For example, men in Blackpool live on average 10 years less than men in Westminster.

Data from Scotland has shown that life expectancy at birth is generally higher, and the all-cause mortality rate is lower, in ethnic minority groups, despite them generally being more deprived than White people.^{[footnote 3]}

We do not yet have data on life expectancy by ethnicity in England, where 97% of the UK’s ethnic minorities live (they are due to be published in May 2021). However, overall age-standardised mortality rates, which are closely correlated with life expectancy, from 2014 to 2018 were lower for the Black and Asian ethnic groups than the national average, according to Public Health England data, again despite higher levels of deprivation.^{[footnote 5]}

Data from 2011 to 2013 in England on mortality by country of birth also showed decreased mortality overall in those aged less than 75, those born in South Asia and Africa, and in women born in the Caribbean (men were not significantly different).

A recent study looking at 2019 data from the Opensafely platform also showed a decreased risk of death (by about a quarter) for Black and South Asian ethnic groups compared with White people. This was also shown in 2020 for non-COVID-19 mortality but there was no overall mortality reduction in 2020 due to the increased death rate from COVID-19. ^{[footnote 34]}

The general pattern shown over the last few years is surprising, as deprivation is the main factor associated with lower life expectancy and higher mortality, and some ethnic minorities (for example, from the Pakistani and Bangladeshi ethnic groups) have better outcomes despite being much poorer than White people (overall). About 30% of Pakistani and Bangladeshi people, and 20% of Black people live in the most deprived 10% of areas, compared with less than 10% for White people and those from the Indian ethnic group.

Healthy life expectancy

The gap between the most and least deprived is even wider for healthy life expectancy where the difference is 20 years on average (33 years between Blackpool and Westminster). ^{[footnote 35]}

Figure 3: Healthy life expectancy at birth by deprivation in England. ^{[footnote 1]}

This deprivation gap or social gradient in life expectancy and healthy life expectancy is likely to be present within all major ethnic groups but this data is not currently available.

There is limited data on healthy life expectancy by ethnicity. One publication from Scotland using linked 2011 Census data showed that despite having longer life expectancy than White women, women from the Pakistani and Indian ethnic groups had shorter healthy life expectancy. ^{[footnote 4]}

Health-related quality of life

There is a discrepancy between more subjective measures of health, where ethnic minority groups (not including White minorities) groups generally do worse, and objective measures. For example, where ethnic minorities have better outcomes for the majority of the leading causes of death and disability, even without adjusting for deprivation. Generally self-reported health is correlated with objective measures, but this correlation is not strong in ethnic minority groups (not including White minorities). ^{[footnote 36]}

Leading causes of mortality

Of the 25 leading causes of mortality as measured by ‘years of life lost’ (YLL), South Asian and Chinese people have better outcomes than White people in more than half of these, while Black people have better outcomes for about a third, worse for about a third and no different for a third.

Cancer

Cancer is the leading cause of death overall in the UK, accounting for 28% of all deaths in the UK. ^{[footnote 37] [footnote 38]}

Of the 25 leading causes of death by YLL, 12 are cancers:

• lung cancer (2nd most common cause of death)
• colorectal cancer (8th)
• breast cancer (9th)
• pancreatic cancer (13th)
• other cancers (16th)
• oesophageal cancer (18th)
• brain Cancer (19th)
• prostate cancer (20th)
• leukaemia (21st)
• stomach cancer (22nd)
• lymphoma (23rd)
• ovarian cancer (24th)

Lung, bowel, breast and prostate cancers together accounted for almost half (45%) of all cancer deaths in the UK in 2017.

Around one-fifth of all cancer deaths are from lung cancer.

1 in 2 people in the UK born after 1960 will be diagnosed with some form of cancer during their lifetime.

Smoking is the largest cause of cancer in the UK, accounting for 15% of all cancer cases.

Figure 4: Mortality for all cancers

Figure 5: Incidence for all cancers

Scotland

Data from Scotland using individual level data linked to the Census has shown that people from Indian and Pakistani ethnic backgrounds have much lower rates of cancer than White people, while people from African and Chinese ethnic backgrounds have slightly lower rates.^{[footnote 39]}

England

A summary of findings based on the main national studies published ^{[footnote 7] [footnote 18] [footnote 19] [footnote 27] [footnote 28] [footnote 29] [footnote 40] [footnote 41]} is given below:

• Compared with White people, people from South Asian ethnic groups have a much lower incidence of ‘all cancers’ as well as every one of the 11 leading causes of cancer deaths. They also have lower mortality for all cancers.

• Black people generally have a slightly lower incidence for all cancers and most of the leading cancers but significantly increased risk of stomach and prostate cancer. They also have lower mortality for all cancers.

• People from the Chinese ethnic group also generally have a slightly lower incidence for all cancers and all of the leading cancers except stomach cancer.

• For all cancers, and 7 out of 11 leading causes of cancer death, White people have the highest incidence and mortality, and in some cases poorer survival.

• These differences are likely to be due to differences in the important risk factors for these cancers (for example, tobacco, alcohol, diet) and the lower rates seen for these cancers in ethnic minority groups (not including White minorities) may give some indication of the potential for prevention in White people.

• People from South Asian ethnic groups do have higher incidence of some less common or fatal cancers, including head and neck, liver, gallbladder, Hodgkin’s lymphoma and thyroid cancer. Black people also have higher rates of cancers of the liver, gallbladder, prostate, uterus, as well as non-Hodgkin lymphoma, thyroid cancer and myeloma.

The different patterns of cancer incidence seen across each of the different ethnic groups as well as differences by sex, age, and cancer subtype, suggest that these findings are unlikely to be due to systematic reporting biases in any of the ethnic groups compared with White people. The increased risks in particular of many cancers in ethnic minority groups supports the absence of an under-reporting bias which has been a concern due to ethnic groups having historically poorer access to healthcare including cancer screening. ^{[footnote 42] [footnote 43] [footnote 44]}.

Where the level of risk factors by ethnicity was known, findings were generally in keeping with what would be expected (for example, smoking tobacco and lung cancer, chewing tobacco, and head and neck cancer, Hepatitis B and liver cancer, HIV and lymphoma, parity and breast cancer) further giving confidence that the other differences in incidence where risk factors are unknown are real. ^{[footnote 7] [footnote 18] [footnote 19] [footnote 27] [footnote 28] [footnote 29] [footnote 40] [footnote 41]}

In South Asian ethnic groups, there was significant heterogeneity of risk for many cancers between the Indian, Pakistani and Bangladeshi ethnic groups, highlighting the importance of analysing them separately. For the Black African and Black Caribbean ethnic groups, this heterogeneity of risk was also apparent for some cancers. This is to be expected for the majority of cancers given the differences between diets, habits and socio-cultural practices in the 3 South Asian groups, and between the Black African and Black Caribbean groups.

For certain cancers, however, the incidence was unusually high or low in all 3 South Asian groups or both Black groups which is suggestive of genetic predisposition (for example, prostate, myeloma, pituitary for Black people, gallbladder and thyroid in people from South Asian ethnic backgrounds) or protection (malignant melanoma in both Black people and people from South Asian backgrounds.) ^{[footnote 7] [footnote 18] [footnote 19] [footnote 27] [footnote 28] [footnote 29] [footnote 40] [footnote 41]}

What explains the disparities in cancer rates?

The lower incidence of many cancers in South Asian ethnic groups, even when the majority of them have spent most of their lives in the UK or were born here, is striking. This contrasts with, for example, the experience of Japanese migrants to the USA who were found to have similar rates of a number of cancers (for example, colorectal) to White American people within one generation ^{[footnote 45]}. This could be due to dietary factors, with most people from South Asian ethnic groups still maintaining a fairly typical South Asian diet, or there may be genetic differences which provide some protection against certain cancers. There may also be potential for cancer prevention if, for example, aspects of the diet are found to be protective. It is also interesting to note that while the incidence of some cancers (for example, lung, breast and colon) are lower for people from South Asian ethnic backgrounds, rates of diabetes and ischaemic heart disease are higher than in White people, even though some of the risk factors are similar ^{[footnote 46]}.

In general, cancer incidence in the South Asian group tended to be closer to that of White people among those aged under 50 years, most of whom were born in the UK or migrated as children ^{[footnote 47]}, than among those older than 50 years (virtually all born outside the UK). ^{[footnote 48]} The notable exception was for breast cancer in under 50s. ^{[footnote 7] [footnote 18] [footnote 19] [footnote 27] [footnote 28] [footnote 29] [footnote 40] [footnote 41]}

This is consistent with environmental exposures, particularly at younger ages, being important in the aetiology of these cancers and it is unlikely that ethnicity itself (or genetic factors) are responsible for most of the observed differences in incidence with ethnicity acting as a proxy for environmental and lifestyle factors (including smoking, chewing tobacco, alcohol and diet) ^{[footnote 49]}

The pattern in Black people was more mixed which may reflect the different patterns of migration for the Black African and Black Caribbean ethnic groups, and so is harder to interpret.

In general, as would be expected for most cancers (where environment is the most important risk factor), the incidence in the migrant population was between the incidence rates in country or region of origin and White people. This would be explained by changes in environment for migrants with the adoption of ‘Western’ habits and lifestyles ^{[footnote 49]} However, there were some notable exceptions with the somewhat unusual finding that the incidence in the ethnic group was higher than both country or region of origin and White people (for example, cancers of the thyroid, prostate, stomach, gallbladder, myeloma, non-Hodgkin lymphoma in Black people and of the thyroid, liver, gallbladder and Hodgkin lymphoma in South Asian people). This is likely to be due to under-diagnosis or under-reporting in many of the countries of origin due to the limited access to healthcare facilities and lack of comprehensive cancer registration.^{[footnote 50]} Also, there may be genetic predisposition to developing these cancers in these ethnic groups which means they maintain high incidence even after migration. ^{[footnote 7] [footnote 18] [footnote 19] [footnote 27] [footnote 28] [footnote 29] [footnote 40] [footnote 41]}

For other cancers, mainly in South Asian ethnic groups, incidence was lower than both country or region of origin and White people (for example, stomach, cervix, malignant melanoma) reflecting a reduction in exposure to the harmful risk factors after migration (for example, reduced exposure to H. pylori, HPV, ultraviolet B radiation). ^{[footnote 7] [footnote 18] [footnote 19] [footnote 27] [footnote 28] [footnote 29] [footnote 40] [footnote 41]}

For the cancers where data is available, survival is generally better or the same for lung, prostate and colorectal cancer in ethnic minority groups with mixed evidence for breast cancer.^{[footnote 7]} This may reflect decreased uptake of screening for breast cancer ^{[footnote 51] [footnote 52]} where people in South Asian ethnic groups and Black people generally have lower uptake of screening, which is also the case for colorectal cancer ^{[footnote 53]} and cervical cancer.^{[footnote 52]}

Cardiometabolic diseases (CVD, stroke and diabetes)

Cardiovascular diseases (CVD), including stroke, are a leading cause of mortality in the UK, contributing to 27% of all deaths in 2019.^{[footnote 54]} Ischaemic heart disease (IHD) is the commonest type of CVD, including myocardial infarction, and heart failure. Cerebrovascular disease is the other major form of CVD, and it includes stroke and transient ischaemic attack. Diabetes increases the risk of CVD almost two-fold.

There is striking variation in CVD risk between South Asian and Black group ethnic groups.^{[footnote 8]} CVD prevalence is higher in South Asian groups, and they develop ischaemic heart disease (IHD) at a younger age than White groups. For CVD incidence, the highest risk is in women from the Pakistani ethnic group, and men in the Bangladeshi group. In contrast, CVD prevalence and incidence are lower among people in the Black African and Black Caribbean groups. Men and women in the Chinese ethnic group also have lower CVD incidence than in the White group. ^{[footnote 9]}

Table 6: Adjusted hazard ratios (95% confidence interval) for cardiovascular disease, by ethnicity and sex

Predictor variables	Model A* (women)	Model A* (men)
Townsend score (per 5 unit increase)§	1.48 (1.46 to 1.51)	1.19 (1.17 to 1.20)
White or not recorded	1	1
Indian	1.32 (1.26 to 1.38)	1.31 (1.26 to 1.36)
Pakistani	1.76 (1.66 to 1.87)	1.62 (1.54 to 1.69)
Bangladeshi	1.33 (1.23 to 1.44)	1.70 (1.61 to 1.79)
Other Asian	1.07 (0.985 to 1.16)	1.03 (0.968 to 1.10)
Black Caribbean	0.836 (0.791 to 0.884)	0.700 (0.663 to 0.738)
Black African	0.660 (0.605 to 0.721)	0.671 (0.623 to 0.722)
Chinese	0.710 (0.612 to 0.823)	0.652 (0.574 to 0.740)
Other	0.836 (0.786 to 0.890)	0.770 (0.729 to 0.814)

Table notes: *Includes chronic kidney disease (stage 4 or 5) fractional polynomial terms for age (age and age−2) and body mass index (BMI−2 and BMI−2ln(BMI)), and interactions with age for body mass index, systolic blood pressure, Townsend score, family history of ischaemic heart disease, treated hypertension, atrial fibrillation, type 1 diabetes, type 2 diabetes, chronic kidney disease, and smoking status. §Interaction with age; hazard ratios evaluated at mean age.

People from South Asian ethnic groups had more IHD, hypertension and diabetes, and Black people had more hypertension and diabetes but lower IHD than White people. ^{[footnote 9]}

Stroke is more common in Black people, who have 1.5 to 2.5 times greater risk of having a stroke than White people. People from the South Asian group also have a risk for stroke about 1.5 times greater than White people, particularly those in the Pakistani and Bangladeshi ethnic groups. In contrast people in the Chinese ethnic group have lower risk of stroke than White people. ^{[footnote 55]} Data from the stroke register in London shows that while stroke incidence has decreased by 40% for White people in the past years, it has not decreased for Black people. ^{[footnote 55]}

The lower age of onset of many of these diseases, for example i) heart failure in people from South Asian ethnic groups (6 years younger) and Black people (9 years younger) and ii) stroke in people from the Black African and Caribbean groups, and South Asian groups (about 10 years younger) than White people may be due to the younger age profile of these groups ^{[footnote 56]} and age specific incidence or prevalence rates need to be calculated to see if there is a genuine difference by age).

Type 2 diabetes

Compared with the majority White population, for minority ethnic groups:

Prevalence is higher: When diagnosed biochemically, type 2 diabetes prevalence is up to 3 to 6 fold higher in South Asian and Black ethnic groups than among White people. Self-reported diabetes prevalence is 3 to 5 fold greater among the Bangladeshi, Pakistani, Indian and Black Caribbean ethnic groups compared with the general population. ^{[footnote 57]} The higher diabetes risk among the South Asian and Black Caribbean groups is shared by people of these ancestries in different world regions.

Incidence is greater: Risk of new-onset type 2 diabetes is 2 to 3 fold higher in the South Asian and Black African and Black Caribbean groups compared with White people. ^{[footnote 58]} Ethnic subgroups were examined in national primary care databases. In women, compared with White women, the relative risk of type 2 diabetes incidence was 6 fold higher in the Bangladeshi group, 3 fold in the Pakistani group, 3 fold in the Indian group, 2.4 fold in the Chinese group, 1.5 fold in the Black Caribbean group, and 1.33 fold higher in women in the Black African group. The elevated risk had a similar pattern among men from ethnic minorities, with the Bangladeshi group with highest risk, and the Black Caribbean group with lowest increase, but still 60% greater than that in White men. ^{[footnote 9]}

Diabetes develops at younger age and at lower body mass index (BMI): The mean age of onset of type 2 diabetes is around 5 to 10 years younger in migrant South Asian people than European adults. For incidence rates equivalent to those at a higher BMI of 30 kg/m² in European men and women, cut-points are lower for the South Asian (25.2 kg/m²) and Black African and Black Caribbean groups (27.2 kg/m²). ^{[footnote 59]}

Both the Black and South Asian ethnic groups have greater risk of type 2 diabetes at a lower obesity threshold than in White people. A BMI threshold of 25 kg/m2 (overweight category in White people) has an equivalent risk of type 2 diabetes at a lower threshold of 23 kg/m2 in the South Asian group, and a threshold of 30 kg/m2 (obese category) is equivalent to 27.5 kg/m2. Similarly the thresholds for central adiposity are also lower in both the Black and South Asian groups. NICE has highlighted lower thresholds of BMI and waist circumference (central adiposity) for preventing ill health and premature death in the Black and South Asian groups in the UK. ^{[footnote 60]}

The risk of diabetes starts early in life: Metabolic abnormalities are present at 9 to 10 years of age. Compared with White children, children from the South Asian group had higher HbA1c, insulin, triglyceride, and C-reactive protein level and lower HDL-cholesterol even after adjusting for differences in adiposity. In the Black African and Black Caribbean groups, children’s levels of HbA1c, insulin and CRP were also raised, but not as high as in the South Asian group. ^{[footnote 61]} The prevalence and incidence of type 2 diabetes in ethnic minority children (aged under 16 years) is markedly greater than in White children. ^{[footnote 62]}

Women have greater risk of gestational diabetes mellitus (GDM): The relative risk of GDM is elevated both in the Black African and Black Caribbean groups, and women from the South Asian group compared with European women. Progression from GDM to diabetes is also greater among women from ethnic minorities (not including White minorities). ^{[footnote 63]}

Compared with White groups with diabetes, people in the South Asian group have a greater risk of developing IHD. In contrast, Black people have lower risk of IHD. Both groups have greater stroke incidence, which is relatively more pronounced in the Black African group. Peripheral vascular disease: compared with White groups, people in the South Asian group have lower peripheral vascular disease, but there was no significant difference between Black and White people. ^{[footnote 64]}

Nephropathy: Ethnic minority groups (not including White minorities) compared with White groups have higher incidence of end-stage renal disease, and a younger age at initiation of dialysis treatment. ^{[footnote 65]} People in the South Asian group may have higher risk of over proteinuria than microalbuminuria, indicating possible faster progression to nephropathy). People from Black ethnic groups have greater risk of ESRF secondary to diabetes. ^{[footnote 66]}

Retinopathy: People in the South Asian group have greater retinopathy risk, and at a younger age, than White people, with higher risk of sight threatening diabetic retinopathy. ^{[footnote 66] [footnote 67]} People from the Black African and Black Caribbean groups also have significantly elevated risk of diabetic retinopathy and sight threatening retinopathy compared with White people. ^{[footnote 68]}

Neuropathy: people from the South Asian group have lower rates of peripheral neuropathy than White people, but may experience more painful neuropathy ^{[footnote 66]}.

Heart failure: Among women with diabetes, Heart failure rates were 27% lower in the Black Caribbean group than in White women, and similar (or not significantly different) for ethnic minority groups. In men, all ethnic groups had similar or lower HF rates, except men from the Bangladeshi group, who had a 14% greater risk than White men. (QResearch primary care database, among whom 25,480 cases of heart failure occurred during follow up.) ^{[footnote 69]}

Lower limb amputation: Ethnic minority groups, including the South Asian and Black groups, and the Chinese group, have lower risk of amputation compared with White people (or ethnicity not recorded). (Analyses were based on QResearch database, including 454,575 patients with diabetes, among whom 4,822 new cases of lower limb amputation occurred during follow-up). ^{[footnote 70]}

Summary

People from the South Asian group have a higher risk of type 2 diabetes, IHD and stroke than White people. People from the Black ethnic groups have a higher risk of type 2 diabetes, and stroke, but lower risk of IHD than White people. There are differences by ethnic subgroups. People from the Chinese ethnic group mostly have lower risk of disease compared with White people, but less data is available.

Table 7: risk of cardiometabolic diseases by ethnicity, compared with the White ethnic group (England, March 2017 to May 2020)

	South Asian	Indian	Pakistani	Bangladeshi	Black	Black African	Black Caribbean	Chinese
IHD or CVD	Higher risk	Higher risk	Higher risk	Higher risk	Lower risk	Lower risk	Lower risk	Lower risk
Stroke	Higher risk				Higher risk	Higher risk	Higher risk
Stroke or TIA		No difference	Higher risk	Higher risk		No difference	No difference	Lower risk
Diabetes	Higher risk	Higher risk	Higher risk	Higher risk	Higher risk	Higher risk	Higher risk
Diabetes complications
IHD	Higher risk				Lower risk
Stroke	Higher risk		Higher risk	Higher risk	Higher risk			Lower risk
Peripheral vascular disease	Lower risk				No difference
Nephropathy	Higher risk				Higher risk
Peripheral neuropathy	Lower risk
Amputation	Lower risk	Lower risk	Lower risk	Lower risk	Lower risk	Lower risk	Lower risk	Lower risk
Retinopathy	Higher risk				Higher risk
Heart failure	No difference	No difference	No difference	No difference	Higher risk (men)	No difference	Lower risk	Lower risk (men)
Blood pressure (mean levels)	Lower risk	Lower risk	Lower risk	Lower risk		Higher risk	Higher risk
Hypertension		No difference	No difference	Higher risk (men), Lower risk (women)		Higher risk (women)	Higher risk	No difference
Treated hypertension (QResearch)		Lower risk	Lower risk	Lower risk		Lower risk	Higher risk	Lower risk
Cholesterol	No difference
Lower HDL-C/higher triglycerides	Higher risk				Lower risk
Emergency admission		Lower risk	Higher risk	Lower risk (men)		Higher risk	Higher risk	Lower risk

Table notes: IHD = ischaemic heart disease, CVD = cardiovascular disease, TIA = transient ischaemic attack

What explains the disparities?

The reasons for the high risk of type 2 diabetes, IHD and stroke in South Asian ethnic groups, and of high diabetes and stroke risk (but not IHD) in Black people, are not completely clear, but may be explained in part by differences in risk factors. Age, sex, genetics and ethnicity are fixed factors, but potentially modifiable factors are critical to understand and manage. Socio-economic factors are also relevant.

Modifiable risk factors

(i) Hypertension

Hypertension affects around 25% of UK adults. At least half of CVD is related to high BP, and it is related with many other health outcomes (for example, chronic kidney disease). The management and treatment of hypertension is every effective, but a substantial proportion (a third) are sub-optimally managed and many (around 40%, about 5.5 million people in England) remain undiagnosed. Data on levels of blood pressure (BP), or the proportion of people with hypertension or its treatment by ethnic group are not typically included in routine health reports, for example, Hypertension prevalence estimates for local populations.

People from the Black African ethnic group have both higher systolic and diastolic BP, while people from the South Asian group have lower systolic and similar diastolic BP compared with White people. ^{[footnote 71]} In the South Asian group, the lower BP levels are most marked in Bangladeshis, and least marked in the Indian group (people from the Pakistani group had intermediate lower levels than White people). There is an inter-generational effect. The lower levels of BP in the South Asian group when compared with White people in the UK are not observed in children in the South Asian group, with levels raised in the Indian and Bangladeshi groups compared with White children. ^{[footnote 72]}

In the QResearch study of national level primary care data, the prevalence of treated hypertension was highest in the Black Caribbean group (14.8%) and lowest in the Chinese group (2.4%). Among White people it was 9.5%, followed by a lower proportion of treated hypertension in the Black African (7.6%), Indian (5.7%), Pakistani (4.3%), and Bangladeshi (4.1%) ethnic groups.^{[footnote 9]} The raised BP in ethnic groups needs specific monitoring and managing.

(ii) Blood cholesterol and lipid levels

Total blood cholesterol is used as a process measure for the prevention and management of CVD. There is a striking contrast in lipid distribution by ethnicity. Total cholesterol is not raised in UK South Asian people relative to White people but they have an adverse lipid profile with low HDL-cholesterol and high triglycerides. Raised levels of lipoprotein (a) and of more atherogenic small, dense LDL cholesterol levels have been noted in SA in research studies. However, people from the Black African and Black Caribbean groups have a favourable lipid profile (higher HDL-cholesterol, low triglycerides than White people).^{[footnote 8]}

(iii) Hyperglycaemia

Non-diabetic hyperglycaemia (HbA1c value between 6.0% and 6.4% in non-diabetic people) is more prevalent in South Asian (14.2%) and Black (13.1%) ethnic groups compared with White, Mixed and Other ethnic groups (10.4%). ^{[footnote 73]} Features common to the Black African, Black Caribbean and South Asian groups include an increased risk of glucose impairment (impaired glucose tolerance) and raised insulin, indicating a greater degree of insulin resistance in both groups.^{[footnote 8]}

(iv) Overweight or obesity

Body Mass Index (BMI) is a commonly used marker of obesity calculated from height and weight as weight in kilograms divided by the square of height in metres. BMI is a marker of body fatness which, in turn, is a risk factor for cardiometabolic diseases. BMI is commonly used in surveys and studies because it is quick and easy to measure height and weight and is a relatively good proxy of body fatness. However, the relationship between BMI, body fatness and cardiometabolic risk is not consistent across ethnic groups. Some ethnic minority groups (not including White minorities) are at similar risk of cardiometabolic diseases at lower BMI than white groups.

People from the South Asian group have greater central adiposity than White people, as measured by the waist circumference or waist-to-hip ratio. There has been some work to develop ethnic-specific BMI adjustments for children and adults in the UK. These have been used in a research context, but are not yet well developed enough that they can be recommended for use in surveillance studies. ^{[footnote 74] [footnote 75] [footnote 76]}

There is good quality surveillance data for children in the National Child Measurement Programme (NCMP) where all children in reception and year 6 have their height and weight measured in school annually. The Active Lives survey collects self-reported height and weight from around 150,000 people per year.

The most recent systematic review on ethnic differences in obesity in the UK was published in 2011. ^{[footnote 77]} This found that Black adults had consistently higher risk of obesity than White adults, and that adults and children from the Chinese group had consistently lower risk of obesity than White people. There were no consistent patterns in the South Asian group relative to White people. The review identified a number of important methodological limitations of existing research in this area.

Few studies explore, and statistically adjust for, potential predictors of obesity among ethnic minority groups, particularly known causes of obesity such as socio-economic position, maternal BMI, physical activity and diet, which makes it difficult to know why any ethnic differences arise.

Use of inconsistent ethnic groupings or aggregated ethnic groups, masking known heterogeneity within large composition groups, for example the South Asian group, and making comparisons between studies difficult, for example comparing the South Asian group with the Indian, Pakistani and Bangladeshi groups

Figure 6: Percentage of children in reception and year 6 identified as obese (BMI≥95th centile, using UK90 reference) in NCMP, 2019 to 2020. ^{[footnote 78]}

While socio-economic position is also consistently associated with obesity, particularly in children, ethnic differences in socio-economic position do not explain the ethnic differences in obesity seen in children. ^{[footnote 79]}

Ethnicity had an independent influence on overweight or obesity risk after adjustment for socio-economic position in the National Child Measurement Programme, England. ^{[footnote 79]}

Children from Asian and Black ethnic groups are more likely to have obesogenic lifestyles than their white peers, but these differences are not explained by deprivation, and will require culturally specific interventions to reduce obesity related health inequalities. ^{[footnote 80]}

Table 8: Percentage of adults identified as overweight or obese in Active Lives Survey (2018 to 2019)

Ethnicity	%	Number of respondents
All	62.3	152,979
Asian	56.2	5,254
Black	73.6	1,634
Chinese	35.3	799
Mixed	57	1,654
White British	63.3	131,104
White Other	58.1	8,262
Other	52.6	1,018

There is a consistent finding that adults, and increasingly children, of Black ethnicity have the highest prevalence of overweight and obesity, and that adults and children of Chinese ethnicity have the lowest. At reception (ages 4 to 5) Black children are 3 times more likely to be obese than children from the Chinese ethnic group (15% compared with 5%). In adults the difference is two-fold (74% compared with 36%).

The effects of acculturation on obesity rates:

In adults, obesity converged to the risk in the majority population following acculturation. People from the Indian and Chinese groups were more likely to be obese in the second generation than the first. Adjusting independently for the mixed patterns of acculturative changes and the uniform upward social mobility in all groups increased the risk of obesity in the second generation. Future research needs to consider generation and trans-cultural identities as a fundamental variable in determining the causes of ethnic health inequalities. ^{[footnote 81]}

Interventions to reduce obesity across the population may still need to be sensitive to ethnic, socio-economic and cultural factors that may make some interventions less likely to be engaged with by some population sub-groups.

Behavioural risk factors

Diet, physical activity, tobacco, smoking and alcohol are all critical factors for cardio-metabolic diseases and cancer.

Tobacco

Surveillance data on smoking in young people is collected biennially in the ‘Smoking, drinking and drug use among young people’ survey of secondary school students aged 11 to 15 years in England.

Questions on self-reported smoking prevalence in people aged 18 years and over are included in the Annual Population Survey. This is a household survey across the UK that captures information across a range of important social and socio-economic topics, including health, from around 320,000 individuals per year.

Smoking in adults

Generally, men from the Bangladeshi group had the highest prevalence of cigarette smoking, and prevalence was consistently low for women in the Indian and Pakistani groups.^{[footnote 57]}

In 2019, 13.9% of adults in England were current smokers. Prevalence of current smoking was highest in individuals of Mixed ethnicity (19.5%) and lowest in the Chinese ethnic group (6.7%). Compared with the benchmark of 13.9% in the population as a whole, people from White and Mixed ethnic backgrounds have higher smoking rates, and people from the Asian, Black and Chinese groups have lower smoking rates. ^{[footnote 82]}

Figure 7: Smoking prevalence in adults by ethnic group in England

Smoking in children

Ethnic differences in smoking in children are different from those in adults. Boys and girls from Black and Asian ethnic backgrounds were most likely to have never smoked. Boys and girls from White and Mixed ethnic backgrounds are most likely to be regular smokers ^{[footnote 83]}. This may indicate evolving changes in ethnic differences in smoking that will be reflected in adults in due course.

Use of smokeless tobacco and shisha

While cigarette smoking remains the most common way to consume tobacco in the UK, use of smokeless tobacco (chewing tobacco, snuff) and shisha or hookah are more common in some ethnic groups. All tobacco use is associated with significant health harms. Smokeless tobacco use is associated with lower risk of lung disease and lung cancer than cigarette smoking, but higher risk of mouth cancer. ^{[footnote 84]} Shisha use is associated with many of the same problems as cigarette smoking including lung cancer, other lung diseases, heart disease and low birth weight. ^{[footnote 85]}

The exclusive focus of national surveys on cigarettes may underestimate total exposure to tobacco in some ethnic groups.

In spring 2019, Action on Smoking and Health (ASH) commissioned YouGov to conduct a UK nationally representative survey of use of smokeless tobacco and shisha by ethnicity. More than 12,000 adults took part. This confirmed that people from the Bangladeshi ethnic group were most likely to have ever tried and regularly use smokeless tobacco.^{[footnote 86]}

Table 9: Percentage of UK adults who have tried, regularly use and never tried smokeless tobacco, by ethnicity, 2019 (YouGov Survey, n=12,393)

Smokeless tobacco use	White	Indian	Bangladeshi	Pakistani	Black African and Black Caribbean	Other and Mixed
Ever tried	12	16	29	21	19	20
Use at least monthly	1	5	12	0	5	3
Never tried	86	80	68	69	75	75

Shisha use is most common for people in the South Asian, Other, and Mixed ethnic groups.

Table 10: Percentage of UK adults who have tried, regularly use and never tried shisha, by ethnicity, 2019 (YouGov Survey, n=12,393)

Shisha use	White	South Asian	Black/African/Caribbean	Other/mixed
Ever tried	10	21	16	29
Use at least annually	2	11	6	7
Use less often	9	11	10	22
Never tried	77	58	64	57

People from the South Asian group are more likely to use non-cigarette forms of tobacco such as smokeless tobacco and shisha and this may mean that overall exposure to tobacco is underestimated in some ethnic groups in national surveys that focus exclusively on cigarette smoking.

Alcohol

As for smoking, self-reported information on alcohol consumption in young people is collected biennially in the ‘Smoking, drinking and drug use among young people’ survey.

Numerous different metrics are used to quantify (harmful) alcohol use in the UK. These include ever drinking alcohol, drinking more than the recommended weekly maximums, hospital admissions for alcohol-specific causes, and deaths from alcohol-specific causes. Social pressures may make people even more reluctant to accurately remember and report their alcohol use than other health-associated behaviours, whereas hospital admissions may be a more ‘objective’ indicator of extreme harm.

Young people in South Asian ethnic groups are least likely to have ever drunk alcohol, with White people most likely to have drunk alcohol. The difference between groups is more than five-fold (52% compared with 10%). The prevalence of recent alcohol consumption is also highest in young White people, and lowest in young people in South Asian groups. In this case the ratio is more than ten-fold (13% compared with 1%). ^{[footnote 83]}

Table 11: Percentage of boys and girls in years 7 to 11 (age 11 to 16 years) who have ever drunk alcohol, by ethnicity (2018)

Ever drunk alcohol	White (%)	Mixed (%)	Asian (%)	Black (%)	Other (%)
Boys	51	36	10	21	30
Girls	52	43	10	24	17
Total	52	40	10	23	25

Table 12: Percentage of boys and girls in years 7 to 11 (age 11 to 16 years) who drank alcohol in the last week, by ethnicity (2018)

Drank alcohol in the last week	White (%)	Mixed (%)	Asian (%)	Black (%)	Other (%)
Boys	12	6	1	4	1
Girls	13	7	1	3	7
Total	13	7	1	3	5

Alcohol use in adults

White British men and women are mostly likely to be drinking at hazardous, harmful or dependent levels, and men and women from the Asian ethnic group least likely. White British men are more than 6 times as likely to be drinking at hazardous, harmful or dependent levels than men from the Asian group (22.6% compared with 3.7%), and White British women are more than 5 times as likely to be drinking at this level than women from the Asian group (14.8% compared 2.6%).^{[footnote 87]}

Alcohol related hospital admissions

Alcohol-specific hospital admissions are disproportionately high in White British and White Irish men and women, and disproportionately low for men and women of many other ethnicities, particularly people from Asian ethnic backgrounds. ^{[footnote 1]}

Table 13: Percentage of all and alcohol-specific hospital admissions by ethnicity and gender (England, 2014 to 2015)

Ethnicity	Men: All admissions (%)	Men: Alcohol specific admissions (%)	Women: All admissions (%)	Women: Alcohol specific admissions (%)
White British	75.4	79.1	74.6	84.5
Other white	3.7	4.7	4.4	3.3
White Irish	0.8	1.1	0.7	0.8
Indian	1.8	2.0	1.9	0.4
Pakistani	1.8	0.5	2.0	0.1
Bangladeshi	0.5	0.1	0.6	0.1
Chinese	0.2	0.0	0.3	0.1
Black African	1.0	0.5	1.4	0.4
Black Caribbean	0.9	0.5	1.0	0.5

Physical activity

The Active Lives Survey collects self-reported physical activity and has shown that White adults are most likely to be active, and people of Asian ethnicity are least likely to be active. Men are consistently as or more likely to be active than women in all ethnic groups. However, gender differences are more pronounced in some groups, for example, in Black people (64.6% compared with 51.3%). ^{[footnote 88]}

Figure 8: Self-reported physical activity by ethnicity and gender in adults aged 16 or more in England, 2019 to 2020, Active Lives Survey ^{[footnote 88]}

Physical activity in children

Overall, White children are most likely to be active and Black children least likely to be active. However, patterns are slightly different for boys and girls as shown in the table with Black boys being much more likely to be active than girls (42.2% compared with 28.1%). ^{[footnote 88]}

Table 14: Self-reported physical activity by ethnicity and gender in children in school years 1 to 11 in England, 2019 to 2020, Active Lives Survey ^{[footnote 88]}

Ethnicity	All: Active	All: Fairly active	All: Inactive	Boys: Active	Boys: Fairly active	Boys: Inactive	Girls: Active	Girls: Fairly active	Girls: Inactive
White British	47.2%	24.4%	28.4%	49.0%	22.6%	28.5%	45.4%	26.5%	28.1%
White Other	48.1%	23.8%	28.1%	51.3%	22.4%	26.3%	44.1%	25.7%	30.2%
Asian	41.7%	22.4%	35.9%	47.0%	20.8%	32.2%	36.4%	24.3%	39.3%
Black	35.5%	24.2%	40.4%	42.2%	23.1%	34.6%	28.1%	25.0%	46.9%
Mixed	42.1%	23.2%	34.8%	39.5%	23.6%	36.9%	44.1%	23.4%	32.5%
Other	38.5%	21.7%	39.8%	41.3%	21.6%	37.1%	34.7%	22.8%	42.6%

No studies to date have used validated or objective measures of physical activity and there may be differential biases in measures used between ethnic groups. ^{[footnote 89]}

Possible reasons for the differences described above have been explored. 2 reviews focused on South Asian ethnic groups and found that this group tends to:

• have a good understanding of the link between physical activity and chronic diseases
• underestimates recommended levels of activity
• perceives higher body weights to be healthier than people of White ethnicity
• fears racial harassment when exercising
• lacks culturally appropriate opportunities for group-based activities, such as mixed-gender classes or classes delivered in English ^{[footnote 90] [footnote 91] [footnote 92]}

One review reported that explicit racism in some professional sports, such as football, may contribute to ethnic differences in wider participation in Black and ethnic minority groups more generally. ^{[footnote 93]}

Diet: variation of dietary intakes by ethnic group in the UK

Suboptimal diet is a critical and leading risk factor for non-communicable disease mortality and morbidity ^{[footnote 94]}. Diet is a complex set of variables, but a systematic way to evaluate diet is to consider nutrients (macronutrients, such as fat, carbohydrate and protein; and micronutrients that include vitamins and minerals), food groups (such as fruit and vegetables, meat, fish, and dairy) and overall diet patterns and diet quality. Such a systematic appraisal of diets by ethnic group is absent in the UK. In fact, there is very limited and patchy information on ethnic minority diets, often with contradictory findings.

The National Diet and Nutrition Survey Rolling Programme (NDNS RP) is a continuous cross-sectional survey representative of the UK population. Though participants from ethnic minorities are included in NDNS, analysis by ethnic group has not been conducted due to limited sample size and concerns or representativeness. For instance, the latest report from NDNS does not include any data by ethnic group. ^{[footnote 95]}

Dietary data by ethnic group was collected in the Health Survey for England, 2004, when an ethnic boost sample was included. This is now outdated as the food landscape and dietary habits have shifted considerably since then. ^{[footnote 57]}

Compared with White people, the diets of people from ethnic minority groups is generally:

• better for: (i) saturated fat (lower intake possibly; but more data needed)
• worse for: (i) vitamin D
• inconclusive or unknown for: (i) fruit and vegetable intake and meeting 5-a-day recommendation, (ii) oily fish, (iii) red and processed meat, (iv) polyunsaturated fat, (v) trans fat, (vi) sugar sweetened beverages, (vii) free sugars intake, (viii) salt consumption, (ix) iron status, (x) dietary patterns or overall diet quality

Table 15: Summary of overall findings on the distribution of healthy and less dietary factors in ethnic minority groups compared with the majority White population in the UK

Diet in ethnic minority groups compared with White people	For dietary factors related to improved health	For dietary factors related to poor health
Convincing evidence	vitamin D – more adversely distributed in ethnic minority groups
Probable or possible		Saturated fat – more favourably distributed in ethnic minority groups
Inconclusive, insufficient or unknown	Fruit and veg quantity, Fruit and veg under 5-a-day, Fruit and veg variety, Oily fish, Polyunsaturated fat, Healthy diet patterns or high diet quality	Red and processed meat, Sugar sweetened beverages, Trans fat, Free sugars, Salt, Iron status (but anaemia likely higher prevalence), Unhealthy diet patterns, or poor or low diet quality

It is vital to understand differences in diet quality between different ethnic minority groups and the majority of White people in the UK, but sparse data is available. Most convincing evidence is for lower levels of vitamin D in the South Asian and Black ethnic groups, but granularity by ethnic subgroups is unavailable. There is some evidence that dietary intakes of saturated fat may be lower (more favourable for cardiovascular risk) in the South Asian and Black groups, but generally all groups exceeded the recommendation to consume less than 10% of energy as saturated fat. For all other relevant and highly important dietary factors there is paucity of adequate data to make meaningful comparisons between ethnic groups.

Diet and nutrition are critical for health, but they exert their influence within the context of overall health behaviours including but not limited to physical activity, alcohol consumption and smoking. It is important to examine combined effects of health behaviours, including diet. Though dietary intake information is very limited by ethnicity in cohort studies that enable such research, one analysis was undertaken comparing White and South Asian groups. It showed that lack of adherence to 4 combined health behaviours (non-smoker, moderate alcohol intake, physically active, frequent fruit and vegetable intake) was linked with 2 to 3 fold increased CVD in both White and South Asian groups. The population attributable fraction (indicating potential for disease prevention) was high in both groups but relatively greater in the South Asian group, suggesting the crucial importance of targeting health behaviours and identifying strategies that will be effective. ^{[footnote 58]}

Recommendations

There is a need to redress the gap in understanding dietary habits by ethnic groups. Investment is needed in research and in national surveys to include adequate minority ethnic participants, with ethnic boost samples from time to time. The extent of misreporting bias by ethnic group is unknown. Availability of digital dietary assessment methods and instruments, coupled with potential use of nutritional biomarkers can offer a way forward to redress the research gap.

Socio-economic factors

There is a well-established link between socio-economic disparities and risk of obesity, type 2 diabetes, cardiovascular disease and many cancers. There is also evidence for socio-economic disparity by ethnicity, and it may therefore follow that some ethnic health differences may reflect underlying socio-economic differences.

Disentangling the effects of socio-economic status and ethnic background is complex. Within each ethnic group there is a social gradient, but this may not explain the greater risk seen for some diseases in some ethnic groups (for example, diabetes in the South Asian group).

For example, in the Whitehall study, there was a strong social gradient with 67% of people in the Black African and Black Caribbean group, 50% of people in the South Asian group, and 19% of White European people in the lower end of the grade structure (clerical and support staff). After adjustment for confounding by socio-economic status, however, there remained 2 to 4 times greater risk of type 2 diabetes and hypertension in both ethnic groups compared with White people. ^{[footnote 96]}

In the Scottish national study, compared with White people with type 2 diabetes, people from the Pakistani ethnic group were at increased risk of CVD, whereas people from the Chinese group were at lower risk, with these differences unexplained by known risk factors, including accounting for deprivation. ^{[footnote 97]}

Among people with type 2 diabetes, both socio-economic status and ethnicity are important determinants of inequality. There was a gradient of worsening glycaemic control across socio-economic groups, with the HbA1c of most deprived (IMD quintile 1 and 2) being higher than that in the least deprived group (IMD quintile 5). Black people had worse glycaemic control (HbA1c +2.36 mmol/mol) than White people, as did people from the Asian group (+1.10 mmol/mol). There was no evidence for interaction between deprivation and ethnicity, except SA individuals in IMD quintile 1 (most deprived) were more likely to have monitoring for HbA1c. ^{[footnote 98]} There were no differences in the prevalence of non-diabetic hyperglycaemia by quintiles of deprivation in England national data. ^{[footnote 73]}

However, lower SES does not explain the lower than expected risks observed for overall mortality and other diseases, for example, most cancers in the South Asians group.

Summary

People from the South Asian ethnic group have greater prevalence of metabolic syndrome (elevated BP, blood sugar, central obesity, abnormal HDL-cholesterol and triglyceride levels), together with insulin resistance. These are associated with their higher risks of diabetes, IHD and stroke. In Black people, the lipid profile is more favourable, and may explain their lower IHD risk, but their risk is elevated for diabetes and stroke. The South Asian group typically have hyperinsulinaemia and an adverse lipid profile, while in the Black African and Black Caribbean groups there is a dissociation between hyperglycaemic and dyslipidaemic effects of insulin resistance, the reasons for which are not clear. Regardless, management with lifestyle behavioural factors is relevant in all groups.

There are some shared features and some disparate features among risk factors for the different rates of type 2 diabetes and its complications, and for CVD (IHD and stroke) by ethnic group. Age, sex, genetics and ethnicity are fixed factors, and genetic or biological effects are relevant, but potentially modifiable factors include behavioural risk factors and socio-economic disparities. Both deprivation and ethnicity have associations with disease and with risk factors for diseases, but there does not seem convincing evidence for an independent effect of social disparity in explaining ethnic differences. However, the interface between SES factors and ethnicity is not well studied or understood. There is a need to conduct analyses that specifically check for interaction and joint effects.

Future research needs to see the investigation of differences between risk in ethnic population sub-groups as the purpose of the research endeavour rather than an issue that needs to be dealt with analytically by restricting attention to European sub-populations only.

Role of genetics and epigenetics in explaining differences between ethnic groups

Ethnicity is comprised of a combination of genetic, cultural and geographical factors, and individuals’ self-reported ethnicity is not necessarily consistent with their genetic ancestry. ^{[footnote 99] [footnote 100]} Furthermore, governmental Census categories may be combinations of races, cultural groupings and aggregations of smaller ethnic groups. ^{[footnote 101]} For example, the West African contribution to individual African-American ancestry averages about 80%, but ranges from approximately 20% to 100%. ^{[footnote 102]}

Approximately 85% of genetic variation between human beings exists within members of the same ethnic group with only 10% to 15% being explained by differences between ethnic groups. ^{[footnote 102] [footnote 103]} That is, there is usually more genetic variation between members of the same group than between groups, especially for people from African ethnic backgrounds).

There are some rare diseases which are markedly more common in certain ethnic groups. These are often related to selective advantages of such mutations or historical genetic bottlenecks within populations. For example, the presence of sickle cell disease confers protection against malaria, making it more common for people from African ethnic backgrounds where malaria is endemic. ^{[footnote 104]} These diseases tend to represent a small fraction of the total burden of disease.

Non-disease related genetic variants have also been identified across ethnic groups. For example, people from East Asian ethnic backgrounds are more likely to be alcohol intolerant than people from European ethnic backgrounds due to an inactive variant of an alcohol processing gene, and people from European ethnic backgrounds are more likely to be able to drink milk into adulthood due to the presence of a variant of the enzyme lactase. ^{[footnote 105] [footnote 106]}

The contribution of genetics to common chronic diseases (including cancer, diabetes and obesity) is modest. Although there are clear ethnic differences in risk for many of these disorders, common genetic variants and polygenic risk scores do not, in general, explain much of those differences. There are, however, some exceptions, for example the higher incidence of prostate cancer among people from the Black ethnic group. ^{[footnote 107]}

Data on genetic variations in common disease across ethnic groups are limited. This may reflect a bias within current research, since genetic studies investigating the role of disease tend to predominantly use people from European ethnic backgrounds.

Although there is more genetic diversity among the Black African ethnic group, the majority of genetic information is derived from small studies, mostly from West Africa. ^{[footnote 109]} While there has been an expansion of data for people from European ethnic backgrounds, there is relative stagnation in other ethnic groups. ^{[footnote 108]} Even in those instances where ethnic minority (not including White minorities) data is available, such as in UK Biobank studies often focus attention on the majority European sub-population only in order to reduce heterogeneity. ^{[footnote 108] [footnote 110]}

Epigenetics

Epigenetic factors are modifications of chromosome structures that can affect gene expression and are a potential explanation for between-ethnic group differences in disease risk. ^{[footnote 102] [footnote 111]} Gene expression varies dramatically between cells of different organs, and across different stages of the life-course. ^{[footnote 112]} These changes are dynamic and reversible, indicating a cellular response to the environment that individuals are exposed to which accumulate with age. ^{[footnote 112] [footnote 113]} During certain periods of life (for example, embryonic development) epigenetic changes may be more likely under the influence of environmental factors (such as food scarcity, diet, temperature, environmental toxins). ^{[footnote 112] [footnote 114] [footnote 115]}

Epigenetic modifications may be the mechanism by which environmental exposures affect biological pathways, and gene expression levels reflect gene-environment interactions over time. ^{[footnote 116]} Studies have reported epigenetic modifications resulting from diverse exposures, including air pollution, psychosocial stress and smoking. ^{[footnote 111] [footnote 117]} These epigenetic changes can be associated with health outcomes across generations. ^{[footnote 111] [footnote 116]} Although there are several types of epigenetic modification, DNA methylation is the most commonly studied. ^{[footnote 117]} There has been relatively little research in exploring the contribution of differences in epigenetic markers to between ethnic group differences in health outcomes. This is part due to the intrinsic challenges of epigenetic population research which, unlike genetic research, is impacted on by the environment and the selection of the study population. ^{[footnote 116]}

There is a paucity of evidence as to whether different ethnic groups have the same epigenetic modifications in response to similar environmental exposures. Smoking-associated DNA methylation sites have been reported to be similar between people from African American ethnic backgrounds and White people ^{[footnote 118]}, but few other exposures have been investigated. Many studies link epigenetics to exposures or outcomes, but do not focus on differences in DNA methylation by ethnic group.

A small number of studies have reported variation in methylation levels and patterns by ethnic group with differences noted, for example, in White and Black people for breast and prostate cancers. ^{[footnote 111] [footnote 117]} However, it is unclear whether these changes reflect true differences between ethnic groups or instead reflect differences between groups in the factors (for example, younger age of onset, environment) that are associated with the epigenetic changes. ^{[footnote 111] [footnote 116]}

Healthcare utilisation, disease management and risk factor control

Diabetes

Disparities in areas of clinical care may lead to higher rates of complications and adverse outcomes for some groups. National standards include monitoring of 8 care processes and 3 treatment targets (to optimise HbA1c, BP and statins) in primary care. A breakdown by ethnic group is not included in routine reporting.

Better risk factor recording and initiation of therapy in ethnic minority groups

There is limited evidence of systematic ethnic inequalities around the time of type 2 diabetes diagnosis. In analysis of CPRD of 180k people, ethnic minority groups (not including White minorities) had fewer consultations compared with white groups in the 12 months prior to diagnosis, but risk factor recording was better than or equivalent to white groups for 9 out of 10 risk factors for people from the South Asian group, and 8 out of 10 risk factors for Black people. Time to initiation of antidiabetic treatment and first risk assessment was faster in ethnic minority groups (not including Whiten minorities) relative to White people, but time to risk factor measurement and diabetes review was slower. All analyses were adjusted for age, sex, deprivation, and clustering by practice. ^{[footnote 118]}

Slower intensification of therapy in ethnic minority groups

There are no ethnic differences in initiation of antidiabetic therapy, but there is evidence of more slow intensification of therapy in Black people and people from the South Asian group than White people, with greater therapeutic inertia following identification of uncontrolled HbA1c (failure to intensify treatment within 12 months of HbA1c >7.5% [58 mmol/mol]). At all stages of treatment intensification, ethnic minority groups (not including White minorities) had fewer HbA1c measurements than white groups. Reasons need to be explored for ethnic disparities in downstream consequences, and may be related to poorer long-term monitoring and control of risk factors in ethnic minority groups (not including White minorities). Initiatives to improve timely and appropriate intensification of diabetes treatment are key to reducing disparities in downstream vascular outcomes in these populations. ^{[footnote 118]}

Disparity in disease monitoring and prescribing

Analysis from the Royal College of General Practitioners Research and Surveillance Centre dataset shows ethnic disparity in diabetes management. Black people were less likely to have annual testing for HbA1c and retinopathy, and had worse HbA1c control than White people. People from Asian ethnic backgrounds were more likely to have monitoring for HbA1c and nephropathy, but less likely for retinopathy and neuropathy than White people. Both groups had lower prescription of newer antidiabetic therapies such as sodium-glucose cotransporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP-1) agonists than White people. People in the most deprived category (IMD quintile 1) were more likely to have monitoring for HbA1c, BP, retinopathy, and neuropathy than predicted by ethnicity and IMD alone. ^{[footnote 119]}

These analyses are indicative but not conclusive. Limitations of data need to be addressed. These include variable quality and completeness of routinely recorded data, including but not limited to ethnicity, and a lack of information on medication adherence.

Cardiovascular disease

There are few studies that investigated ethnic and social disparities for CVD event management or procedures. Amid the limited evidence, mostly, there was no evidence for systematic differences by ethnicity. For example, there was no ethnic (or gender) disparity in ‘symptom-to-door times’ in patients presenting with ST elevation myocardial infarction. ^{[footnote 120]}

The Whitehall Study found no evidence that low social position or South Asian ethnicity was associated with lower use of cardiac procedures or drugs, independently of clinical need, and differences in medical care were unlikely to contribute to social or ethnic differences in ischaemic heart disease in this cohort. ^{[footnote 121]}

Use of NHS stop smoking services

Smokers from the Asian ethnic group were most likely to have stayed quit 28 days after quitting (59% of Bangladeshi smokers), whilst successful quit attempts were least common in the Mixed and Black ethnic groups. ^{[footnote 122] [footnote 123]}

Table 16: Number and percentage of people using NHS stop smoking services who remained quit 28 days after their quit date (self-reported), April 2018 to March 2019

		Number setting a quit date	Number of successful quitters	Percentage successful (%)
All ethnic groups		236,175	123,800	52
White		203,244	106,679	52
Asian or Asian British		10,024	5,619	56
Indian		2,860	1,576	55
Pakistani		3,010	1,661	55
Bangladeshi		2,228	1,316	59
Black or Black British		4,772	2,419	51
Caribbean		1,979	978	49
African		1,899	984	52
Chinese		272	155	57

There are some ethnic differences in (self-reported) success of NHS stop smoking services by ethnic group which may contribute to and exacerbate differences in smoking prevalence.

Wider indicators of health-system performance

There are differences in quality indicators of health-system performance across ethnic groups.

Hospital admissions

People from South Asian ethnic groups in Scotland had higher avoidable hospital admissions than the White Scottish group, with the highest rate in the Pakistani ethnic group for both men and women. There may be issues of access to and quality of primary care to prevent avoidable hospital admissions, especially for the South Asian group. There was little variation between ethnic groups in hospital length of stay or unplanned readmission. ^{[footnote 123]}

Table 17: Healthcare utilisation by ethnicity compared with the White ethnic group

Healthcare utilisation	South Asian ethnic group	Indian ethnic group	Pakistani ethnic group	Black ethnic group	Black African ethnic group	Black Caribbean ethnic group	Chinese ethnic group
Healthcare presentation (cardiovascular) [footnote 124]	Significantly better	N/A	N/A	No significant difference	N/A	N/A	N/A
Primary management [footnote 124]	Significantly better	N/A	N/A	No significant difference	N/A	N/A	N/A
Access to specialist cardiovascular care [footnote 124]	Significantly better	N/A	N/A	No significant difference	N/A	N/A	N/A
Avoidable hospital admissions*[footnote 123]	N/A	Significantly worse (S)	Significantly worse (S)	No significant difference (S)	N/A	N/A	Significantly better (S)
Avoidable deaths* [footnote 123]	N/A	Significantly better (S)	Significantly better (S)	N/A	No significant difference (S)	No significant difference (S)	Significantly better (S)
Unplanned readmissions*[footnote 123]	N/A	Significantly worse (S)	Significantly worse (S)	N/A	No significant difference (S)	No significant difference (S)	Significantly better (S)

Table notes – (S): Scotland data only, N/A: data not available

NHS Health Check

Attendance vs non-attendance was slightly better for people from South Asian ethnic groups and Black people compared with White people. ^{[footnote 125]}

Figure 9: Attendance of NHS health checks, by ethnicity

Some conclusions from the data presented

In contrast to the commonly presented view that ethnic minorities (not including White minorities) have universally worse health outcomes compared with White people, the picture is much more mixed.

Of the 25 leading causes of mortality as measured by Years of Life Lost, people from the South Asian and Chinese ethnic groups have better outcomes than White people in more than half of these, while Black people have better outcomes for about a third, worse for about a third and with no difference for a third.

Ethnic minorities have better health outcomes for many of these diseases despite higher levels of deprivation. Life expectancy and overall mortality is generally better in ethnic minorities, but not always healthy life expectancy.

To reduce inequalities in life expectancy and overall mortality, interventions need to focus on their impact on those with the worst outcomes – which is often most closely associated with geography and deprivation, not ethnicity.

White people should not be viewed as a homogenous group as the biggest differences in outcomes are within the White group (for example, the highest and lowest life expectancy due to deprivation).

This also applies to ethnic minorities where there are differences in many health outcomes between individual groups, for example the Indian, Pakistani and Bangladeshi ethnic groups for many cancers. There are also likely to be differences within individual ethnic groups by deprivation but no data is currently available on this.

Despite ethnic minorities faring better on some objective measures of health outcomes, this is not always seen in self-rated quality of life measures.

Most differences in cancer and cardiometabolic diseases outcomes can be explained by differences in known risk factors (exceptions include prostate cancer in Black people and diabetes in people from the South Asian ethnic group) and where data is available, survival is generally either better or no worse for ethnic minorities.

As deprivation is a major risk factor for these outcomes, and it differs significantly by ethnicity, comparisons should also be made after adjusting for deprivation.

---

The following footnotes had to be removed from this version because they accompanied figures and charts we could not reproduce here: ^{[footnote 33]}

1. England, P.H., Public Health Outcomes Framework: Health Equity Report Focus on Ethnicity 2017, Public Health England. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹

2. Unit, R.D., Ethnicity facts and figures. 2021. ↩

3. Gruer, L., et al., Life expectancy of different ethnic groups using death records linked to population Census data for 4.62 million people in Scotland. Journal of Epidemiology and Community Health, 2016. 70(12): p. 1251-1254. ↩ ↩² ↩³ ↩⁴

4. Cezard, G.I., Evidencing the gap between health expectancy and life expectancy for ethnic groups in Scotland. Revue Quetelet Journal 2019. Vol 7 No 1 (2019). ↩ ↩² ↩³

5. England, P.H., Disparities in the risk and outcomes of COVID-19. 2020. ↩ ↩² ↩³ ↩⁴

6. England, P.H., The Burden of Disease in England compared with 22 peer countries: A report for NHS England. 2020. ↩ ↩²

7. Forman, D., Cancer Incidence and Survival by Major Ethnic Group, England, 2002 - 2006. 2009, National Cancer Intelligence Network. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰

8. Chaturvedi, N., ETHNIC DIFFERENCES IN CARDIOVASCULAR DISEASE. Heart, 2003. 89(6): p. 681-686. ↩ ↩² ↩³ ↩⁴ ↩⁵

9. Lawson, C.A., et al., Risk Factors for Heart Failure: 20-Year Population-Based Trends by Sex, socio-economic Status, and Ethnicity. Circ Heart Fail, 2020. 13(2): p. e006472. ↩

10. Hippisley-Cox, J., C. Coupland, and P. Brindle, Derivation and validation of QStroke score for predicting risk of ischaemic stroke in primary care and comparison with other risk scores: a prospective open cohort study. BMJ: British Medical Journal, 2013. 346: p. f2573. ↩

11. Bhopal, R.S., et al., Ethnic variations in the incidence and mortality of stroke in the Scottish Health and Ethnicity Linkage Study of 4.65 million people. Eur J Prev Cardiol, 2012. 19(6): p. 1503-8. ↩

12. Gilkes, A., et al., Does COPD risk vary by ethnicity? A retrospective cross-sectional study. Int J Chron Obstruct Pulmon Dis, 2016. 11: p. 739-46. ↩

13. Bhopal, R., et al., Risk of respiratory hospitalization and death, readmission and subsequent mortality: scottish health and ethnicity linkage study. Eur J Public Health, 2015. 25(5): p. 769-74. ↩

14. Pham, T.M., et al., Trends in dementia diagnosis rates in UK ethnic groups: analysis of UK primary care data. Clin Epidemiol, 2018. 10: p. 949-960. ↩

15. Simpson, C.R., et al., Ethnic variations in morbidity and mortality from lower respiratory tract infections: a retrospective cohort study. J R Soc Med, 2015. 108(10): p. 406-17. ↩

16. Al-Sharifi, A., C.R. Krynicki, and R. Upthegrove, Self-harm and ethnicity: A systematic review. International Journal of Social Psychiatry, 2015. 61(6): p. 600-612. ↩

17. Ali, R., et al., Incidence of gastrointestinal cancers by ethnic group in England, 2001-2007. Gut, 2013. 62(12): p. 1692-703. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

18. Shirley, M.H., et al., Incidence of breast and gynaecological cancers by ethnic group in England, 2001-2007: a descriptive study. BMC Cancer, 2014. 14: p. 979. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

19. Knowles, R.L., et al., Ethnic and socio-economic variation in incidence of congenital heart defects. Arch Dis Child, 2017. 102(6): p. 496-502. ↩

20. Bhopal, R.S., et al., Ethnic variations in heart failure: Scottish Health and Ethnicity Linkage Study (SHELS). Heart, 2012. 98(6): p. 468-73. ↩

21. Steinbach, R., et al., ‘Race’ or place? Explaining ethnic variations in childhood pedestrian injury rates in London. Health Place, 2010. 16(1): p. 34-42. ↩

22. Cézard, G., et al., Ethnic variations in falls and road traffic injuries resulting in hospitalisation or death in Scotland: the Scottish Health and Ethnicity Linkage Study. Public Health, 2020. 182: p. 32-38. ↩

23. NatCen Social Research and the Department of Health Sciences, U.o.L., Mental Health and Wellbeing in England: Adult Psychiatric Morbidity Survey 2014. 2016. ↩

24. Bhala, N., et al., Alcohol-related and hepatocellular cancer deaths by country of birth in England and Wales: analysis of mortality and Census data. J Public Health, 2009. 31(2): p. 250-257. ↩ ↩²

25. Bhala, N., et al., Ethnic Variations in Liver- and Alcohol-Related Disease Hospitalisations and Mortality: The Scottish Health and Ethnicity Linkage Study. Alcohol Alcohol, 2016. 51(5): p. 593-601. ↩ ↩²

26. Maile, E.J., et al., Nervous System and Intracranial Tumour Incidence by Ethnicity in England, 2001-2007: A Descriptive Epidemiological Study. PLoS One, 2016. 11(5): p. e0154347. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

27. Maruthappu, M., et al., Incidence of prostate and urological cancers in England by ethnic group, 2001-2007: a descriptive study. BMC Cancer, 2015. 15: p. 753. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

28. Shirley, M.H., et al., Incidence of haematological malignancies by ethnic group in England, 2001-7. Br J Haematol, 2013. 163(4): p. 465-77. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

29. Steel, N., et al., Changes in health in the countries of the UK and 150 English Local Authority areas 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet, 2018. 392(10158): p. 1647-1661. ↩

30. Fecht, D., et al., Associations between air pollution and socio-economic characteristics, ethnicity and age profile of neighbourhoods in England and the Netherlands. Environ Pollut, 2015. 198: p. 201-10. ↩

31. Statistics, O.f.N., Life expectancy for local areas of the UK: between 2001 to 2003 and 2017 to 2019. 2020. ↩

32. Bhaskaran, K., et al., Factors associated with deaths due to COVID-19 versus other causes: population-based cohort analysis of UK primary care data and linked national death registrations within the OpenSAFELY platform. medRxiv, 2021: p. 2021.01.15.21249756. ↩

33. Statistics, O.f.N., Health state life expectancies by national deprivation deciles, England: 2016 to 2018. 2020. ↩

34. Watkinson, R.E., M. Sutton, and A.J. Turner, Ethnic inequalities in health-related quality of life among older adults in England: secondary analysis of a national cross-sectional survey. The Lancet Public Health. ↩

35. Foundation, B.H., https://www.bhf.org.uk/what-we-do/our-research/heart-statistics/heart-statistics-publications/cardiovascular-disease-statistics-2020. 2020. ↩

36. UK, C.R., Cancer mortality for all cancers combined. 2020. ↩

37. Bhopal, R.S., et al., Does the ‘Scottish effect’ apply to all ethnic groups? All-cancer, lung, colorectal, breast and prostate cancer in the Scottish Health and Ethnicity Linkage Cohort Study. BMJ Open, 2012. 2(5). ↩

38. Ali, R., et al., Cancer incidence in British Indians and British whites in Leicester, 2001-2006. British Journal of Cancer, 2010. 103(1): p. 143-148. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

39. Finlayson, A., et al., Incidence of thyroid cancer in England by ethnic group, 2001-2007. Br J Cancer, 2014. 110(5): p. 1322-7. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

40. Access to health care and minority ethnic groups. 2006, King’s Fund: London. ↩

41. Szczepura, A., C. Price, and A. Gumber, Breast and bowel cancer screening uptake patterns over 15 years for UK south Asian ethnic minority populations, corrected for differences in socio-demographic characteristics. BMC Public Health, 2008. 8(1): p. 346. ↩

42. Webb, R., et al., Uptake for cervical screening by ethnicity and place-of-birth: a population-based cross-sectional study. J Public Health (Oxf), 2004. 26(3): p. 293-6. ↩

43. McCracken, M., et al., Cancer Incidence, Mortality, and Associated Risk Factors Among Asian Americans of Chinese, Filipino, Vietnamese, Korean, and Japanese Ethnicities. CA Cancer J Clin, 2007. 57(4): p. 190-205. ↩

44. Forouhi, N., et al., Do known risk factors explain the higher coronary heart disease mortality in South Asian compared with European men? Prospective follow-up of the Southall and Brent studies, UK. Diabetologia, 2006. 49(11): p. 2580-2588. ↩

45. Swerdlow, A.J., et al., Cancer mortality in Indian and British ethnic immigrants from the Indian subcontinent to England and Wales. Br J Cancer, 1995. 72(5): p. 1312-9. ↩

46. Office for National Statistics, UK Census 2001. Available at: http://www.ons.gov.uk/ons/guide-method/census/census-2001/index.html, London: Office for National Statistics. ↩

47. Parkin, D.M. and M. Khlat, Studies of cancer in migrants: rationale and methodology. Eur J Cancer, 1996. 32A(5): p. 761-71. ↩ ↩²

48. Ferlay J, S.I., Ervik M, Forman D, Bray F, Dikshit R, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11. 2012; Available from: http://globocan.iarc.fr/ia/World/atlas.html. ↩

49. Bansal, N., et al., Major ethnic group differences in breast cancer screening uptake in Scotland are not extinguished by adjustment for indices of geographical residence, area deprivation, long-term illness and education. Br J Cancer, 2012. 106(8): p. 1361-6. ↩

50. Moser, K., J. Patnick, and V. Beral, Inequalities in reported use of breast and cervical screening in Great Britain: analysis of cross sectional survey data. BMJ, 2009. 338. ↩ ↩²

51. Campbell, C., et al., Are there ethnic and religious variations in uptake of bowel cancer screening? A retrospective cohort study among 1.7 million people in Scotland. BMJ Open, 2020. 10(10): p. e037011. ↩

52. Foundation, B.H., Heart and Circulatory Disease Statistics 2020. 2020. ↩

53. Hippisley-Cox, J., C. Coupland, and P. Brindle, Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study. BMJ, 2017. 357: p. j2099. ↩ ↩² ↩³ ↩⁴

54. Wang, Y., A.G. Rudd, and C.D. Wolfe, Age and ethnic disparities in incidence of stroke over time: the South London Stroke Register. Stroke, 2013. 44(12): p. 3298-304. ↩ ↩²

55. figures, E.f.a., Age profile by ethnicity - England and Wales. 2020. ↩

56. Sproston, K.a.M.J., Health Survey for England 2004: The Health of Minority Ethnic Groups 2006. ↩ ↩² ↩³

57. Tillin, T., et al., Insulin resistance and truncal obesity as important determinants of the greater incidence of diabetes in Indian Asians and African Caribbeans compared with Europeans: the Southall And Brent REvisited (SABRE) cohort. Diabetes Care, 2013. 36(2): p. 383-93. ↩ ↩²

58. Tillin, T., et al., Ethnicity-specific obesity cut-points in the development of Type 2 diabetes - a prospective study including three ethnic groups in the United Kingdom. Diabet Med, 2015. 32(2): p. 226-34. ↩

59. Excellence, N.I.f.H.a.C., BMI: preventing ill health and premature death in black, Asian and other minority ethnic groups. 2013. ↩

60. Whincup, P.H., et al., Early emergence of ethnic differences in type 2 diabetes precursors in the UK: the Child Heart and Health Study in England (CHASE Study). PLoS Med, 2010. 7(4): p. e1000263. ↩

61. Khanolkar, A.R., et al., Ethnic Minorities Are at Greater Risk for Childhood-Onset Type 2 Diabetes and Poorer Glycemic Control in England and Wales. Journal of Adolescent Health, 2016. 59(3): p. 354-361. ↩

62. Rayanagoudar, G., et al., Quantification of the type 2 diabetes risk in women with gestational diabetes: a systematic review and meta-analysis of 95,750 women. Diabetologia, 2016. 59(7): p. 1403-1411. ↩

63. Sebastianski, M., et al., Paradoxically lower prevalence of peripheral arterial disease in South Asians: a systematic review and meta-analysis. Heart, 2014. 100(2): p. 100. ↩

64. Owusu Adjah, E.S., et al., Prevalence and incidence of complications at diagnosis of T2DM and during follow-up by BMI and ethnicity: a matched case–control analysis. Cardiovascular Diabetology, 2018. 17(1): p. 70. ↩

65. Gupta, R. and A. Misra, Epidemiology of microvascular complications of diabetes in South Asians and comparison with other ethnicities. J Diabetes, 2016. 8(4): p. 470-82. ↩ ↩² ↩³

66. Fischbacher, C.M., et al., Is there equity of service delivery and intermediate outcomes in South Asians with type 2 diabetes? Analysis of DARTS database and summary of UK publications. Journal of Public Health, 2009. 31(2): p. 239-249. ↩

67. Sivaprasad, S., et al., Ethnic variations in the prevalence of diabetic retinopathy in people with diabetes attending screening in the United Kingdom (DRIVE UK). PLoS One, 2012. 7(3): p. e32182. ↩

68. Hippisley-Cox, J. and C. Coupland, Development and validation of risk prediction equations to estimate future risk of heart failure in patients with diabetes: a prospective cohort study. BMJ Open, 2015. 5(9): p. e008503. ↩

69. Hippisley-Cox, J. and C. Coupland, Development and validation of risk prediction equations to estimate future risk of blindness and lower limb amputation in patients with diabetes: cohort study. BMJ : British Medical Journal, 2015. 351: p. h5441. ↩

70. Modesti, P.A., et al., Panethnic Differences in Blood Pressure in Europe: A Systematic Review and Meta-Analysis. PLoS One, 2016. 11(1): p. e0147601. ↩

71. Battu, H.S., R. Bhopal, and C. Agyemang, Heterogeneity in blood pressure in UK Bangladeshi, Indian and Pakistani, compared with White, populations: divergence of adults and children. Journal of Human Hypertension, 2018. 32(11): p. 725-744. ↩

72. England, P.H., NHS Diabetes Prevention Programme (NHS DPP) Non-diabetic hyperglycaemia, N.C.I.N. (NCVIN), Editor. 2015. ↩ ↩²

73. Hudda, M.T., et al., Body mass index adjustments to increase the validity of body fatness assessment in UK Black African and South Asian children. International Journal of Obesity, 2017. 41(7): p. 1048-1055. ↩

74. Ntuk, U.E., et al., Ethnic-Specific Obesity Cutoffs for Diabetes Risk: Cross-sectional Study of 490,288 UK Biobank Participants. Diabetes Care, 2014. 37(9): p. 2500. ↩

75. Hudda, M.T., et al., Patterns of childhood body mass index (BMI), overweight and obesity in South Asian and black participants in the English National child measurement programme: effect of applying BMI adjustments standardising for ethnic differences in BMI-body fatness associations. International Journal of Obesity, 2018. 42(4): p. 662-670. ↩

76. El-Sayed, A.M., P. Scarborough, and S. Galea, Ethnic inequalities in obesity among children and adults in the UK: a systematic review of the literature. Obesity Reviews, 2011. 12(5): p. e516-e534. ↩

77. NHS, National Child Measurement Programme, England 2019/20 School Year. 2020. ↩

78. Strugnell, C., et al., Variation in the socio-economic Gradient of Obesity by Ethnicity – England’s National Child Measurement Programme. Obesity, 2020. 28(10): p. 1951-1963. ↩ ↩²

79. Falconer, C.L., et al., Can the relationship between ethnicity and obesity-related behaviours among school-aged children be explained by deprivation? A cross-sectional study. BMJ Open, 2014. 4(1): p. e003949. ↩

80. Smith, N.R., Y.J. Kelly, and J.Y. Nazroo, The effects of acculturation on obesity rates in ethnic minorities in England: evidence from the Health Survey for England. European Journal of Public Health, 2011. 22(4): p. 508-513. ↩

81. England, P.H., Local Tobacco Control Profiles. 2020. ↩

82. Digital, N., Smoking, Drinking and Drug Use among Young People in England 2018 [NS]. 2019. ↩ ↩²

83. Moles, D.R., et al., Oral and pharyngeal cancer in South Asians and non-South Asians in relation to socio-economic deprivation in South East England. British journal of cancer, 2008. 98(3): p. 633-635. ↩

84. Akl, E.A., et al., The effects of waterpipe tobacco smoking on health outcomes: a systematic review. Int J Epidemiol, 2010. 39(3): p. 834-57. ↩

85. Health, A.o.S.a., Tobacco and Ethnic Minorities. 2019. ↩

86. figures, E.f.a., Harmful and probable dependent drinking in adults. 2018. ↩

87. England, S., Active Lives Survey. 2020. ↩ ↩² ↩³ ↩⁴

88. Fischbacher, C.M., S. Hunt, and L. Alexander, How physically active are South Asians in the United Kingdom? A literature review. Journal of Public Health, 2004. 26(3): p. 250-258. ↩

89. Patel, N., et al., Barriers and Facilitators to Healthy Lifestyle Changes in Minority Ethnic Populations in the UK: a Narrative Review. Journal of Racial and Ethnic Health Disparities, 2017. 4(6): p. 1107-1119. ↩

90. Ige-Elegbede, J., et al., Barriers and facilitators of physical activity among adults and older adults from Black and Minority Ethnic groups in the UK: A systematic review of qualitative studies. Preventive Medicine Reports, 2019. 15: p. 100952. ↩

91. Teijlingen, S.K.P.S.E.v., Review of Barriers to Engaging Black and Minority Ethnic Groups in Physical Activity in the United Kingdom. Home Journals Global Journal of Health Science Archives Vol. 1, No. 2 (2009) Koshoedo, 2009. ↩

92. sport, U., Systematic Review of the Literature on Black and Minority Ethnic Communities in sport and physical recreation. 2009. ↩

93. Health effects of dietary risks in 195 countries, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet, 2019. 393(10184): p. 1958-1972. ↩

94. England, P.H., NDNS: results from years 9 to 11 (combined) – statistical summary. 2020. ↩

95. Whitty, C.J., et al., Differences in biological risk factors for cardiovascular disease between three ethnic groups in the Whitehall II study. Atherosclerosis, 1999. 142(2): p. 279-86. ↩

96. Malik, M.O., et al., Ethnicity and risk of cardiovascular disease (CVD): 4.8 year follow-up of patients with type 2 diabetes living in Scotland. Diabetologia, 2015. 58(4): p. 716-25. ↩

97. Ajetunmobi, O., et al., Informing the ‘early years’ agenda in Scotland: understanding infant feeding patterns using linked datasets. J Epidemiol Community Health, 2014. 68(1): p. 83-92. ↩

98. Kaseniit, K.E., et al., Genetic ancestry analysis on >93,000 individuals undergoing expanded carrier screening reveals limitations of ethnicity-based medical guidelines. Genet Med, 2020. 22(10): p. 1694-1702. ↩

99. Census, race and science. Nat Genet, 2000. 24(2): p. 97-8. ↩

100. Anderson NB, B.R., Cohen B, editors., Critical Perspectives on Racial and Ethnic Differences in Health in Late Life. 2004. ↩

101. Bamshad, M., Genetic influences on health: does race matter? Jama, 2005. 294(8): p. 937-46. ↩ ↩² ↩³

102. Pearce, N., et al., Genetics, race, ethnicity, and health. Bmj, 2004. 328(7447): p. 1070-2. ↩

103. Hussein, N., et al., Preconception risk assessment for thalassaemia, sickle cell disease, cystic fibrosis and Tay-Sachs disease. Cochrane Database Syst Rev, 2018. 3(3): p. Cd010849. ↩

104. Edenberg, H.J., The genetics of alcohol metabolism: role of alcohol dehydrogenase and aldehyde dehydrogenase variants. Alcohol Res Health, 2007. 30(1): p. 5-13. ↩

105. Hollox, E., Evolutionary genetics: genetics of lactase persistence–fresh lessons in the history of milk drinking. Eur J Hum Genet, 2005. 13(3): p. 267-9. ↩

106. Zeigler-Johnson, C.M., et al., Genetic susceptibility to prostate cancer in men of African descent: implications for global disparities in incidence and outcomes. Can J Urol, 2008. 15(1): p. 3872-82. ↩

107. Chan, A.T., et al., Long-term use of aspirin and nonsteroidal anti-inflammatory drugs and risk of colorectal cancer. JAMA, 2005. 294(8): p. 914-23. ↩

108. Bien, S.A., et al., The Future of Genomic Studies Must Be Globally Representative: Perspectives from PAGE. Annu Rev Genomics Hum Genet, 2019. 20: p. 181-200. ↩ ↩²

109. Sudlow, C., et al., UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med, 2015. 12(3): p. e1001779. ↩

110. Ahmad, A., et al., Epigenetic basis of cancer health disparities: Looking beyond genetic differences. Biochim Biophys Acta Rev Cancer, 2017. 1868(1): p. 16-28. ↩ ↩² ↩³ ↩⁴ ↩⁵

111. Cavalli, G. and E. Heard, Advances in epigenetics link genetics to the environment and disease. Nature, 2019. 571(7766): p. 489-499. ↩ ↩² ↩³

112. Handy, D.E., R. Castro, and J. Loscalzo, Epigenetic modifications: basic mechanisms and role in cardiovascular disease. Circulation, 2011. 123(19): p. 2145-56. ↩

113. Matboli, M., et al., Emerging role of nutrition and the non-coding landscape in type 2 diabetes mellitus: A review of literature. Gene, 2018. 675: p. 54-61. ↩

114. Brown, W.M., Exercise-associated DNA methylation change in skeletal muscle and the importance of imprinted genes: a bioinformatics meta-analysis. Br J Sports Med, 2015. 49(24): p. 1567-78. ↩

115. Ng, J.H., et al., Explaining the Relationship Between Minority Group Status and Health Disparities: A Review of Selected Concepts. Health Equity, 2019. 3(1): p. 47-60. ↩ ↩² ↩³ ↩⁴

116. Vick, A.D. and H.H. Burris, Epigenetics and Health Disparities. Curr Epidemiol Rep, 2017. 4(1): p. 31-37. ↩ ↩² ↩³

117. Mathur, R., et al., Ethnic disparities in initiation and intensification of diabetes treatment in adults with type 2 diabetes in the UK, 1990-2017: A cohort study. PLoS Med, 2020. 17(5): p. e1003106. ↩ ↩² ↩³

118. Whyte, M.B., et al., Disparities in glycaemic control, monitoring, and treatment of type 2 diabetes in England: A retrospective cohort analysis. PLoS Med, 2019. 16(10): p. e1002942. ↩

119. Brown, R.A., et al., Symptom-to-door times in patients presenting with ST elevation myocardial infarction–do ethnic or gender differences exist? Qjm, 2016. 109(3): p. 175-80. ↩

120. Britton, A., et al., Does access to cardiac investigation and treatment contribute to social and ethnic differences in coronary heart disease? Whitehall II prospective cohort study. Bmj, 2004. 329(7461): p. 318. ↩

121. Digital, N., Statistics on NHS Stop Smoking Services in England April 2018 to March 201. 2019. ↩

122. Katikireddi, S.V., et al., Assessment of health care, hospital admissions, and mortality by ethnicity: population-based cohort study of health-system performance in Scotland. The Lancet Public Health, 2018. 3(5): p. e226-e236. ↩ ↩² ↩³ ↩⁴ ↩⁵

123. Asthana, S., et al., Inequity in cardiovascular care in the English National Health Service (NHS): a scoping review of the literature. Health and Social Care in the Community, 2018. 26(3): p. 259-272. ↩ ↩² ↩³

124. programme, N.H.C., Patients attendance for NHS Health Check by ethnicity 2017-18. 2020. ↩

125. Bhopal, R.S., et al., Mortality, ethnicity, and country of birth on a national scale, 2001-2013: A retrospective cohort (Scottish Health and Ethnicity Linkage Study). PLoS Med, 2018. 15(3): p. e1002515. ↩