Premature mortality and ethnicity

Interpreting linked Census and death-registration evidence on premature mortality in England

The Office for National Statistics (ONS) has linked 2021 Census ethnicity to death registrations. With support from The Health Foundation, our new report looks at premature mortality rates across ethnic groups in England, and how these vary across wider determinants of health.

The data shows that most Black, Asian and minoritised ethnic groups have lower premature mortality than White British people. Our report asks how that pattern should be understood.

The report is published alongside analysis from The Health Foundation of the same data, which has a focus on socioeconomic inequalities.

Explore the report

Swipe through our carousel for a short guide to the report.

It explains how the COVID-19 crisis exposed gaps in ethnicity data, how the new ONS linkage changes that, and why the headline pattern is not so simple.

The carousel ends with our key takeaway on what an effective response to premature mortality must include.

Cover slide with a dark blue X-ray background of a chest and spine. White text reads “Premature Mortality and Ethnicity: What does the new data tell us?” Race Equality Foundation logo, “Supported by The Health Foundation” logo, and a white heartbeat line. Text: “New report from the Race Equality Foundation, 10 November. Underpinned by modelling from the Office for National Statistics.”
Slide with the heading “The COVID-19 crisis exposed serious gaps in ethnicity data.” Text explains that ethnicity is not routinely recorded on death registrations, which made it harder to identify the scale of unequal vulnerability, illness and death, despite Black, Asian and other minoritised ethnic communities raising concerns early in the pandemic. A grid of blue squares and an X-ray strip decorate the slide.
Slide with the heading “But new linked data now offer a fuller picture.” Text explains that the ONS has linked ethnicity recorded in the 2021 Census to death registrations, allowing premature mortality to be examined alongside people’s socioeconomic circumstances. The report looks at premature mortality (dying before 75) across ethnic groups in England by deprivation and place of birth, and is published alongside a Health Foundation report on the same dataset. Decorative blue squares and a coloured bar chart illustration.
Slide headed “The headline pattern: Most minoritised ethnic groups have lower age-adjusted premature mortality than White British people. But what does this mean? The data is not so simple. Swipe to learn more.” Below are two bar charts of age-adjusted premature mortality rates, for women (orange-red) and men (purple). Among women, rates range from 181 for Chinese to 370 for White British. Among men, they range from 284 for Chinese to 573 for Black Caribbean, with White British men at 537. Men’s rates are higher than women’s in every group.
Slide with the heading “The pattern also differs by:” followed by three blue labels with icons: sex (female and male symbols), country of birth (globe) and deprivation levels (pound sign). Text below reads: “Lower observed mortality in several groups does not mean that ethnic health inequalities are absent, or that racism is irrelevant.” A blue heartbeat line runs along the bottom.
Slide with the heading “Taken together, the findings challenge any simple account of ethnicity and premature mortality.” Text: “Our report places them alongside wider evidence showing how health is shaped by social and economic conditions, including education, employment, housing and income.” Four panels with icons show Education (graduation cap), Employment (briefcase), Housing (house) and Income (pound coin), each with a white heartbeat line.
Slide with the heading “The ONS analysis does not measure racism directly.” Text explains that wider evidence shows determinants of health are not separate from racism, as their unequal distribution is often shaped by structural racism, and that racism can therefore affect health both directly and through the conditions in which people live and work. A diagram shows “Racism” with two arrows, “Direct racism” and “Indirect racism” (with education, employment, housing and income icons), both leading to “Effects on Health Outcomes”. The outcome is shown over a circular photo of a man lying with his eyes closed and a hand on his chest.
Key takeaway slide: “An effective response to premature mortality requires both population-wide action on the wider determinants of health and targeted action to tackle racial inequality and racism.” Text: “Read the full report and explore the findings on our website.” Race Equality Foundation and Health Foundation logos, with the web address raceequalityfoundation.org.uk/health-care/premature-mortality.

Key findings

  • 1
    Most ethnic groups have lower age-adjusted premature mortality than White British people.
  • 2
    People born outside the UK generally have lower observed rates than UK-born people of the same ethnicity.
  • 3
    Deprivation gradients differ by ethnic group; the gradient is steepest for White British people.
  • 4
    Racism works directly and indirectly, through education, employment, income, housing, neighbourhoods and healthcare.
  • 5
    Adjusting for wider determinants of health doesn’t remove racism from the analysis, because some of these may be pathways for racism.
  • 6
    Action on race equality must include the wider determinants of health.
  • 7
    Action is needed in deprived White British communities, without minimising racism or setting populations against one another.
  • 8
    Ethnicity, country of birth and the wider determinants are interrelated, not competing explanations.

Report figures

Drawing on linked 2021 Census and death-registration data, the graphs below show:

  • Figure 1: Premature mortality by ethnicity and sex
  • Figure 2: Premature mortality by ethnicity, sex and country of birth
  • Figure 3: Premature mortality across five deprivation quintiles, by ethnicity and sex
  • Figure 4: Absolute premature mortality across ethnicity and socioeconomic domains
  • Figure 5: Age-adjusted and fully adjusted ratios by ethnicity and sex

Further Resources

Office for National Statistics dataset

ONS logo

 

 

Read More

The Health Foundation report

THF Logo

Read More

Explore our work further