Research and analysis

Health inequalities in health protection report 2025

Updated 14 May 2025

Executive summary

Health inequalities in health protection have a high human cost across people and place. They have a wider societal impact, including on health services and economic productivity. The causes of and solutions to addressing health inequalities are often systemic, structural and complex. This report sets out the extent of these health inequalities. It aims to support health systems in better understanding who and where is most affected. It is a step towards enabling sustainable change, so that all our communities can live longer and in better health, safe from health hazards.

In England, those living in the 20% most deprived areas bear the greatest burden, with emergency hospital admission rates due to infectious disease almost twice as high as compared to the least deprived. People from more deprived areas are also disproportionately impacted by radiation, chemical, climate and environmental hazards through their exposure, direct impacts on their health, and the exacerbation of existing health conditions.

Poor outcomes from infectious diseases are unevenly distributed across ethnic groups, with some ethnic minority groups experiencing the highest emergency hospital admissions. Emergency hospital admission rates for tuberculosis (TB) were 29 times higher for the ‘Asian other’ group, 27 times higher for the Indian group and 15 times higher for the ‘Black African’ group, compared to the ‘White British’ group. For all infectious diseases considered together, 5 ethnic group categories had rates of emergency hospital admission disproportionately higher than the ‘White British’ group. In-depth analysis is required to understand the multiple contributing factors to these inequalities and the interaction between demographic characteristics. People from some ethnic minority groups are also disproportionately impacted by radiation, chemical, climate and environmental hazards.

Across England regions, those living in the North West bore the greatest burden of infectious disease, with overall emergency hospital admission rates in the region 1.3 times higher than the England average and over 1.5 higher than the South East.

Inclusion health groups, such as people seeking asylum, people in prison, people experiencing homelessness and people who inject drugs are often disproportionately impacted by a range of infectious diseases. For example, it is estimated that over 80% of people in England living with chronic Hepatitis C have an injecting drug history. However, inclusion health groups are often not visible in routine health surveillance data.

As well as the cost to the physical and mental health of our communities, these inequalities are costly to the health system; inequalities in infectious disease emergency hospital admissions are estimated to amount to between £970 million and £1.5 billion per year in avoidable costs to the NHS.

The findings of this report are a starting point: more work is needed to understand the causes for these disparities and there will be many confounding factors within the data. Applying a health equity lens to our work in health protection, including through the data and evidence, can catalyse action in reducing inequalities through prevention and early treatment in priority areas and support in achieving more equitable outcomes. This includes:

• improving vaccine outcomes
• progressing World Health Organization (WHO) elimination goals on bloodborne viruses and HIV
• managing outbreaks and incidents
• supporting partners across national and local government to mitigate impacts on health services, such as impacts from winter pressures and from the wider determinants of health

Background and purpose

Inequalities in health protection outcomes are unfair, avoidable, persistent and pervasive across people and place. This includes health inequalities in infectious disease and infections, and in exposure and vulnerability to environmental hazards. As well as the physical and mental impacts that infectious diseases and infections have on the health of communities, there is also a financial burden for health systems. Inequalities in infectious disease emergency hospital admissions between the most and least deprived areas in England are estimated to amount to between £970 million and £1.5 billion (calculations in this report are taken from the financial year 6 April 2022 to 5 April 2023). Inequalities in infection, hospital admissions, and mortality are a significant cost to the NHS. They directly contribute to the gap between the richest and poorest in life expectancy and healthy life expectancy. Moreover, they place a disproportionate burden on certain groups and communities.

UKHSA is committed to striving for equitable health outcomes in health protection, as set out in both the UKHSA Strategic Plan and our Health Equity for Health Security Strategy. To achieve this, we need to understand where inequalities are greatest, for whom and for which hazards (including infectious diseases and other hazards in UKHSA’s remit). We also need to understand the interventions that best mitigate inequalities experienced by different groups.

This report therefore aims to provide a high-level description of the current state of inequalities in health protection in England. We focus on 3 dimensions of inequality:

• people living in different areas of deprivation
• people from different ethnic groups
• geographical inequalities (England, at regional level and Integrated Care System level as of April 2025)

The focus on these populations aligns with the UKHSA CORE20PLUS framework adapted from NHS England. We recognise that there are some caveats with the data, which we have set out in the report. Understanding differences will support targeted and evidence-based strategies and interventions to be developed at local, regional and national levels to reduce inequalities and improve health security across England.

This report primarily uses emergency hospital admission rates for infectious diseases as a proxy for infectious disease burden. This will under-represent the true burden of infectious disease, because it does not capture the full extent of infections or illness in the community, only people admitted to hospital. Non-severe cases of infection can be under-reported. Nevertheless, hospital admissions can provide useful insight into the frequency and severity of infections and the allocation of healthcare resources, and tend to be reasonably accurately coded by disease. Admission rates can be influenced by a number of factors including:

• pre-existing health conditions (1)
• vaccination uptake (2, 3)
• place of birth
• risk of infection
• severity of illness
• differences in health-seeking behaviour (4)
• access to local healthcare and preventative services
• hospital admission thresholds

There are also data limitations for demographic information. For example, ethnicity data may not always be reliably recorded. Further analysis and research is required to fully understand the underlying confounding and contributing factors to explain these observed differences. Notwithstanding these limitations, the patterns and trends in this report can highlight variation in infectious disease burden and be used as a catalyst to explore the factors behind these differences.  

Notwithstanding these limitations, the patterns and trends in this report can highlight variation in infectious disease burden and be used as a catalyst to explore the factors behind these differences. This report adds new analysis on inequalities for all emergency hospital admissions for infectious diseases and infections. This builds on existing surveillance, which often considers specific infections in isolation, or is not able to disaggregate by deprivation, ethnic group, or geography. We use current data, consistent across infectious diseases, which allows us to make novel comparisons between types of infectious disease and infection. This helps us to target interventions and to better understand the patterns of current inequalities. For environmental hazards, we worked with subject matter experts across UKHSA to summarise trends from existing evidence on inequalities by deprivation, ethnic group and region.

This work therefore acts as a starting point for future analyses on health inequalities in health protection. It is a starting point for us and for others to delve deeper into more granular geographies and communities. We also need to work more closely in partnership with communities to identify how best to tackle hazard-specific inequalities. Finally, we need to address the contributing factors to inequalities in health protection outcomes to drive sustainable change.

Infectious diseases

Deprivation

People living in more deprived areas have worse health outcomes due to infectious disease than people in less deprived areas.

In England, for the year 1 September 2023 to 31 August 2024, hospital admission rates due to infectious disease and infection were nearly twice as high for people in the 20% most deprived areas (deprivation quintile 1: “IMD1”), as compared to the least deprived (deprivation quintile 5: “IMD5”). Respiratory infections had the greatest inequality, with an estimated additional 260,000 admissions due to inequalities associated with deprivation, in total (calculated as if all levels of deprivation had the same admission rates as compared to the least deprived quintile: IMD1, IMD2, IMD3, IMD4 as compared to IMD5). If comparing the most deprived quintile (IMD1) to the least deprived quintile (IMD5), there were 128,000 admissions attributable to inequality (Figure 1).

Figure 1. Emergency admission rates from infectious diseases and infections, by deprivation quintile, England, 1 September 2023 to 31 August 2024 (age- and sex-standardised rates)

The legend is presented in the same order as the stacked bars.​

Sources: Secondary Uses Service Admitted Patient Care (emergency admission), ONS 2021 Census (population denominators).

Figure 1 shows how emergency admission rates were twice as high for respiratory diseases, twice as high for invasive infections and 1.7 times higher for gastrointestinal diseases, for those living in the most deprived areas compared to the least. Within disease categories, inequality in admissions rates between people living in the most and least deprived areas is almost universal, however, the scale of inequalities varies by condition. Figure 2 provides a selection to illustrate the variability. In this analysis, emergency admission rates were 7 times higher for tuberculosis (TB), 6 times higher for measles and 3 times higher for influenza, for those living in the most deprived areas compared to the least. These disease specific rate ratios are subject to important limitations and caveats described below Figure 2.

Figure 2. Inequalities in emergency hospital admissions between the most deprived and least deprived quintiles, England, 1 September 2023 to 31 August 2024  (age- and sex-standardised rates) [Note 1]

Note 1: The reference bubble of value 1.0 shows the size if rates were equal for the most and least deprived, 2.0 shows the size if rates are twice as high for the most deprived areas compared to the least deprived areas, and so on. The degree of shading represents the total number of emergency admissions recorded in England for the condition for the most deprived quintile, IMD1 . Limitations of these rate ratios are given below and in the ‘Further Information’ section.

Figure 2 uses reference bubbles to show the ratios for admissions of different diseases in relation to levels of deprivation .

This analysis only considers admissions to hospital. There are some infections that rarely cause hospital admission but may also have marked inequalities, for example some sexually transmitted infections and bloodborne viruses. These are not included in Figure 2 (5). Analyses are based on lists of diagnostic codes recorded in hospital data sets following admissions, rather than laboratory confirmed data or physiological markers of infection (particularly important for sepsis identification). These code lists can sometimes be inaccurate and identify that an infection has occurred when it has not, or miss infections that have occurred.

Accuracy can vary by disease type and infection prevalence which differs by patient groups. For example, for measles, recent analysis suggests around a third of suspected cases that are coded as measles may turn out not to be measles when laboratory tests are completed. In time periods when measles is rare this proportion may be higher. 

For sepsis, the accuracy of sepsis ICD-10 codes in identifying ‘true’ sepsis is highly variable over time, across patient groups (with typically greater accuracy in intensive care unit (ICU) populations than non-ICU populations) and across geographies. This is in part due to differences in codes used to define sepsis, methods in developing algorithms, and the appropriate coding of patients using ICD-10 sepsis codes. It is unknown in the UK to what extent ICD-10 codes under or overestimate ‘true’ sepsis cases, due to changes in coding guidance and awareness of sepsis. The UKHSA Sepsis Surveillance team are working with partners to develop pragmatic sepsis surveillance. The team has established an expert clinical coding group to develop a Suspicion of Sepsis ICD-10 code list that does not rely on sepsis codes, will be less subject to sepsis coding bias and will allow for comparisons over time, across patients and geographies.

These limitations and others are further described in the ‘Further Information’ section below. While they are unlikely to account for the extensive inequalities observed in the report, they mean that numbers should be taken as indicative of the scale of inequalities, rather than as precise point estimates.

Ethnicity

Some ethnic minority groups consistently experienced worse health outcomes due to infectious disease.

In England 2023 to 2024, for all infectious disease admissions considered together, there were 8 ethnic groups out of 15 with admission rates higher than the ‘White British’ group. Of these, 5 ethnic groups had rates that were 1.25 times higher or more, compared to the ‘White British’ group, and these differences were statistically significant. Based on guidance from the Race Disparity Unit, this size of difference is a high priority for policy action, because the admission rates are disproportionate (Figure 3).

There were 7 ethnic group categories that had lower rates of admission compared to the ‘White British’ group. Of these, 4 groups had rates 0.8 times lower or less as compared to the ‘White British’ group (the ‘Chinese’ group, the ‘Mixed White’ and ‘Asian’ group, the ‘White Irish’ group and the ‘Black Caribbean’ group). These were disproportionately lower. The infectious disease category with the greatest absolute inequality between ethnic groups, on average, was respiratory admissions. Emergency admission rates for TB were 29 times higher for the ‘Asian other’ group, 27 times higher for the ‘Indian’ group and 15 times higher for the ‘Black African’’ group, compared to the ‘White British’ group (Table 1).

The infectious disease category with the greatest absolute inequality (rate difference) between ethnic groups, most commonly, was respiratory admissions. Within this category, relative inequality (rate ratios) were highest for tuberculosis (TB). Emergency admission rates for TB were 29 times higher for the ‘Asian other’ group, 27 times higher for the ‘Indian’ group and 15 times higher for the ‘Black African’ group, compared to the ‘White British’ group (Table 1).

Figure 3. Emergency admission rates from infectious diseases and infections by ethnic group category, England, 1 September 2023 to 31 August 2024 (age- and sex- standardised rates, per 100,000) [Note 2]

The legend is presented in the same order as the stacked bars​.

Sources: Secondary Uses Service Admitted Patient Care (admissions), ONS 2021 Census (population denominators).

Note 2: Across all ethnic group categories, we have not accounted for differences between non-UK born and UK-born individuals.

Figure 3 is a stacked bar chart showing the emergency admissions rates by ethnic group category for infectious disease. The results demonstrate the inequalities in admission by ethnic group outlined in the paragraphs above.

Within disease categories, the scale of inequalities between ethnic groups varies by specific disease. Table 1 (below) provides a selection to illustrate the variability. In this analysis, there were 13 ethnic group categories with disproportionately higher admission rates for TB. There were 9 ethnic groups with disproportionately higher admission rates for influenza and 8 ethnic group categories with disproportionately higher admission rates for skin and soft tissue infections. This is compared to the ‘White British’ group admission rate.

Recognising that admission numbers are low, emergency admission rates for TB were 41 times higher for the Black other group, 29 times higher for the Asian other group and 27 times higher for the Indian group, compared to the White British group. Emergency admission rates for skin and soft tissue infections were 3 times higher for the Black other group, 2 times higher for the Pakistani group and 2 times higher for the other ethnic group, compared to the White British group (Table 1).

Table 1. Inequalities between ethnic minority groups and the White British group in emergency hospital admissions, England, 1 September 2023 to 31 August 2024 (age and sex-standardised rates) [Note 3]

Note 3: Refer to the further information section for detailed limitations for this analysis. This analysis presents the data, but there will be a number of underlying contributing and confounding factors to explain these observed differences, beyond the scope of this report. There are also some important limitations to note for the analyses examining specific diseases and infections in Table 1. These limitations are shared in common with Figure 2 above, with further detail in the Further Information section below.

Outbreaks of rare vaccine preventable infections can be highly localised and when these occur in areas where more people within a specified ethnic group live, this may inflate rate ratios. In TB for example, ethnicity differences may be due to risk of exposure in country of origin in those born, or travelling from, outside of the UK.

Some infections that rarely cause hospital admission are not included here as they are outside the scope of the current analysis. These have important inequalities between ethnic groups, for example, for sexually transmitted infections and bloodborne viruses and future work will look into this in more detail.

Using consistent methodology across different infection groupings allows some comparison of which infections show the most marked inequalities. However, we should not be over reliant on the precise point estimates for rate ratios in specific groups, rather they give an indication of the potential magnitude of inequalities to support work to effect change.

Geography

Hospital admission rates are disproportionate between geographies at Integrated Care System (ICS) and NHS region level.

At ICS level, in the year 1 September 2023 to 31 August 2024, Northamptonshire had disproportionately higher age- and sex-standardised admission rates as compared to the England average, followed by South Yorkshire, North East and North Cumbria, Birmingham and Solihull and Black Country. The ICS with the lowest admission rate was Dorset, followed by Cornwall and the Isles of Scilly, North Central London, Frimley, Mid and South Essex. Within these geographies there will be intra-regional variation and inequalities, and further analysis is required to understand the extent of these.

At NHS region level in the year 1 September 2023 to 31 August 2024, the North West region had the highest overall rate of emergency admissions for infectious disease and infections (Figure 4). In relative terms, people from the North West region were 1.3 times more likely to be admitted to hospital for an infectious disease compared to the England population. This is a disproportionate difference.

Figure 4. Emergency admission rates from infectious diseases and infections, per 100,000 by NHS Region and Integrated Care System level (ICS), 1 September 2023 to 31 August 2024 (age- and sex- standardised rates)

Sources: Secondary Uses Service Admitted Patient Care (admissions), ONS 2022 Health Region Populations (population denominators).

Figure 4 shows a map of England split into regions and integrated care systems. It uses darker shades of blue to show areas with higher emergency admission rates (such as Northamptonshire) and lighter shades for areas with lower admissions (such as Dorset).

The respiratory hazard category had the greatest absolute inequality across the 7 England regions (Figure 5). In the North West, where the absolute number of emergency admissions was the greatest, respiratory hazards accounted for approximately an additional 1,700 admissions per 100,000 compared to the England population. Respiratory admissions are also much higher in the North compared to the South overall. Additionally, the North West region consistently had the highest relative inequality for most hazard categories (with some individual ICS areas with higher rates). For instance, people from the North West region were 1.5 times more likely to be admitted to hospital for invasive infections, compared to the England population; this was a disproportionate difference.

Figure 5 (below) shows 7 maps of England split into regions. It uses shading to show differences in emergency admissions rates for 7 different types of infections. These infection types are:

• contact
• gastrointestinal
• invasive infections
• other communicable disease
• other infections
• respiratory
• sexually transmitted infections (STIs), bloodborne viruses (BBVs) or HIV

Figure 5. Emergency hospital admission rates by region and infection type, NHS England regions, 1 September 2023 to 31 August 2024 (age- and sex-standardised rates)

Sources: Secondary Uses Service Admitted Patient Care (admissions), ONS 2022 Health Region Populations (population denominators).

Environmental hazards

Deprivation

Extreme temperatures result in increased mortality. There were nearly 2,300 estimated heat-related deaths (6) during summer 2023, and an estimated 5,500 cold-related deaths during winter 2022 to 2023. Outdoor air pollution (7) is associated with an estimated 29,000 to 43,000 deaths per year in the UK and a wide range of adverse health effects. Exposure to indoor air pollutants (8) was linked to respiratory, neurological and cardiovascular health effects as well as cancers. It can be difficult to track inequalities in health outcomes due to radiation, chemical, climate and environmental hazards. There is often a lack of robust data, it is difficult to attribute the effects of hazards on health, and the risk of morbidity and mortality is often related to existing vulnerabilities as well as exposure to a hazard.

Areas of high levels of deprivation typically have higher levels of air pollution than less deprived and less ethnically diverse areas (9). Individuals living in more deprived areas are also more exposed to adverse weather through varied routes, including thermally inefficient housing, fuel poverty, a lack of air conditioning and distance from cooling environmental greenery (10, 11). Those in the most deprived income quintile are at greatest risk of temperature-related hospital attendance and mortality (10, 12, 13). For every 1°C increase in temperature above a threshold of 16°C (the comparator) there is a 1.02% increase in all-cause attendance amongst the most deprived (13).

Those living in deprived areas are also known to be at greater risk from flooding and to be less likely to have insurance cover and other forms of protection to reduce the impact of a flooding event (14). Flooding has measurable impacts on mental health, with prevalence of mental ill-health higher in those without insurance (more likely for those living in deprivation) (15).

People living in the most deprived areas also have a higher risk of exposure to chemicals, for instance, unintentional non-fire related carbon monoxide mortalities (16), lead exposure (17) and non-toxicological impacts, such as odour from landfills (18). However, the relationship between deprivation and chemical exposure is complex and needs to be considered on an individual chemical basis. The relationship between noise exposure and deprivation varies at a macro level across regions and at a micro level within geographical output areas (19).

Ethnicity

The research literature on ethnicity-related risks from adverse weather exposure in the UK is limited. However, levels of deprivation are higher among most ethnic minority groups than in the general population. Ethnic minority groups are often concentrated in urban areas and are at high risk of poor health outcomes from the urban heat island effect and fluvial flooding (20). Rates of many of the long-term health conditions which predispose individuals to greater health risk from adverse weather exposures are also known to be higher among certain ethnic groups in the UK – including cardiovascular disease, diabetes and multi-morbidity (21, 22). For example, South Asian ethnic groups have higher rates and worse outcomes from cardiovascular disease (23), and both South Asian and Black ethnic groups have higher rates of type 2 diabetes (24), making these groups theoretically more vulnerable to extreme temperatures (14, 19).

Additionally, areas of high ethnic diversity typically have higher levels of air pollution than less deprived and less ethnically diverse areas (9). People from certain ethnic minority groups are also disproportionately represented among those living in overcrowded housing conditions. These conditions can amplify their vulnerability to exacerbated conditions due to environmental hazards (21). Multigenerational housing is also known to be higher in ethnic minority groups which, alongside overcrowded conditions, could be considered an environmental risk factor. This is because it increases the risk of infection and more severe outcomes (25).

Geography

Areas of high ethnic diversity and high levels of deprivation typically have higher levels of air pollution than less deprived and less ethnically diverse areas (9). Inequitable health outcomes are likely to arise due to affordability of housing (availability, location and quality), co-exposures (including interdependencies with indoor air quality and temperature via ventilation), co-morbidities and coping capacities. All of these will have links with deprivation.

In 2023, the London region (administrative geographies) had the highest air pollution, with 1.2 times the concentration of fine particulate matter, as compared to England overall. The fraction of mortality attributable to particulate air pollution was also highest in London at 6.2% in 2023, compared to 5.2% for England overall (26). There is considerable variation within London itself (27).

There are also spatial variations in exposure and health burden attributable to noise across England (28). However, based on existing surveillance, there are no significant inequalities in heat- and cold-related mortality between regions (21).

Figure 6. Table and bar chart showing the fraction of mortality attributable to particulate air pollution, 2023 proportion – %

Area	Value
England	5.2
London region (statistical)	6.2
East Midlands region (statistical)	5.6
East of England region (statistical)	5.4
West Midlands region (statistical)	5.1
Yorkshire and the Humber (statistical)	5.1
South East region (statistical)	5.1
North West region (statistical)	4.9
North East region (statistical)	4.6
South West region (statistical)	4.3

Note 4: Data taken from the DHSC fingertips dashboard.

Inclusion health groups

Inclusion health groups are disproportionately impacted by infectious disease, but the true burden is difficult to quantify, primarily because of limited data.

People belonging to inclusion health groups, such as people seeking asylum, people in prison, people experiencing homelessness, and people who inject drugs, were also disproportionately impacted by a range of infectious diseases. In 2021, in England, TB notification rates are highest in people born outside of the UK, with the rate in non-UK born people at 37.6 (per 100,000) compared to 2.1 in people born in the UK. In 2021, estimated TB incidence rates (per 100,000) were 50.4 in asylum seekers entering the UK since 2018, 30.2 in people experiencing homelessness, and 28.1 in people in prison; the incidence for England overall was 7.8 (29).

People who inject drugs are at disproportionately increased risk of bloodborne viruses. Of the 55,900 adults in England estimated to be living with chronic Hepatitis C (HCV) infection in 2023, 84.6% are estimated to have a current or past drug injecting history (30). People who inject drugs are also at disproportionately high risk of invasive bacterial infections, including bacteraemia due to MRSA, MSSA and Invasive Group A streptococcal infections (31).

In screening exercises conducted amongst people experiencing homelessness during the COVID-19 pandemic, uptake of screening was 63.6%, of which 17.7% were found to be HCV antibody (Ab) positive and 10.5% were HCV RNA positive, indicating current infection. Prevalence was also higher amongst those recorded as a past or current injecting drug user (44.5% HCV Ab positive and 23.6% HCV RNA positive) (32).

In 2022, 72 out of 76 (94.7%) cases of diphtheria in England were in recently arrived asylum seekers, many of whom arrived from areas of conflict where access to immunisation was poor (33).

Gypsy and traveller sites are often located close to pollutants from road traffic, sewage works, recycling centres and industrial pollution (34). People experiencing homelessness are also likely to have increased exposure to road traffic pollution and to the effects of severe weather events.

Detailed information on infections in prisoners will be included in a forthcoming report by the Chief Medical Officer.

However, inclusion health groups are not visible in most infectious disease surveillance data sources. Additionally, there are substantial uncertainties in the size of inclusion health populations making comparison of infection rates challenging. This means that inferences about disease risk often need to be derived from published research. For example, international evidence from high income countries suggests that mortality rates from infectious diseases in people from important inclusion health groups including people who inject drugs, people experiencing homelessness and sex workers are around 11-fold higher than the general population (35).

Economic cost estimates

Infectious diseases and infections have a physical and mental impact on the health of individuals, families and communities. However, there is also an economic cost. The cost of hospital admissions is one metric which can be used to estimate part of the cost of all health inequalities to the health system, not limited to inequalities in health protection alone.

For the financial year 2022 to 2023, inequalities in the cost of NHS infectious disease emergency admissions between deprivation quintiles were estimated at over £970 million per year for infectious disease or infection as a primary diagnosis. Where infectious disease or infection were a primary or secondary diagnosis (including up to four secondary diagnoses), inequalities in the cost between deprivation quintiles were up to £1.5 billion per year (Figure 7). Therefore, the cost of these admissions is estimated to be between £970 million and £1.5 billion per year. A 2009 analysis estimated that in addition to being costly to health services, health inequalities are estimated to cost the UK between £31 billion and £33 billion in lost productivity and between £28 billion and £32 billion in lost tax revenue (36).

Figure 7. Total cost of emergency hospital admissions for infectious disease and infections, by deprivation level, diagnosis fields 1 to 5, England, 6 April 2022 to 5 April 2023

Note 5: Admissions data from Hospital Episode Statistics for the financial year 6 April 2022 to 5 April 2023, the National Tariff Payment System or the National Cost Collection are used to estimate costs of each admission. This is calculated as the excess cost for IMD quintiles 1 to 4 above the cost if all groups had the same age-standardised admission rates as IMD5. Costs are not calculated as the cost attributable to inequality but are the observed difference in costs between levels of deprivation. Costs attributable to inequality also depend on severity of admission, length of stay and mortality, which are not included in this analysis.

Figure 7 is a stacked bar chart showing the total cost of infections disease and infections emergency hospital admissions by deprivation level. Costs are calculated per patient and aggregated by deprivation level. The legend is presented in the same order as the stacked bars which represent the level of diagnosis fields (from primary to fourth secondary).

UKHSA’s approach to reducing health inequalities in health protection

Achieving more equitable health protection outcomes is a key goal of UKHSA’s corporate strategy. By interrogating the data to identify inequalities and their root causes, UKHSA can develop tailored health protection advice, guidance, and system leadership to achieve health security for every person in every community. We do this by partnering with communities, academia, commercial partners, local and national government, and the health family to co-create population specific and place-based approaches.

Infectious disease

UKHSA undertakes analysis, modelling and use of behavioural and health services science to develop interventions and guidance. These support the effective prioritisation and roll out of preventative services such as vaccination, testing and treatment for infectious diseases, incident response and pandemic preparedness. This includes developing tailored interventions specifically for inclusion health populations, who can face some of the most severe forms of disadvantage and multiple barriers to care leading to high levels of infection, emergency hospitalisation and poor outcomes.

We work across government, the NHS, with academic partners, with industry and communities themselves to improve the health of populations who experience inequalities. UKHSA works with these groups to understand their needs and co-develop approaches to improve access to care and tailor public health messaging to meet these. Our tailored approaches to prevention, testing and treatment contribute to the UK commitments to the WHO elimination goals on BBV and HIV and support in mitigating impacts on health services, such as from winter pressures; and inform the design and commissioning of efficient and effective health services, such as for inclusion health groups.

When outbreaks occur, UKHSA leads and delivers expert health protection services locally and nationally to respond to cases and outbreaks and prepare for scenarios such as pandemics. Taking lessons from COVID-19 and more recent incidents, we aim to systematically embed health equity considerations throughout all we do, so that we start with recognising the diverse nature and needs of our population and build that into all of our preventative and response work. We provide expertise to support cross government responses to health incidents together with local partners, the voluntary and charity sector, to prevent the spread of infectious disease and reduce the impact of health inequalities. Some examples include advising the Home Office on health protection risks relating to shared accommodation for asylum seekers; providing advice to reduce outbreaks of measles and associated operational challenges in prison settings, and leading communication and control efforts across government to support mpox response in a range of settings and populations.

Environmental hazards

Working with Devolved Administrations, UKHSA provides public health leadership across chemical, radiological and nuclear risks, leading the health protection response to these incidents.

UKHSA is building the evidence base regarding effective interventions to protect health from hazards, such as heat waves and flooding, and is monitoring the impact of climate change on these hazards, such as in the Health Effects of Climate Change report. This includes identifying how inequalities are exacerbated by environmental hazards, and effective interventions to mitigate these and increase community resilience, for example, developing guidance to mitigate the impact of extreme health events for people experiencing homelessness (37).

UKHSA is also supporting work across government to address the wider determinants of health that will contribute to existing gaps in healthy life expectancy, such as the health impacts of outdoor and indoor air quality, and effective interventions to reduce these.

Next steps

UKHSA aims to protect every person in every community in England and support the government’s goal to improve people’s chances of living well for longer. Further information regarding how UKHSA delivers equitable outcomes and supports wider government efforts to address health inequalities is available in the UKHSA’s Health Equity for Health Security Strategy UKHSA is currently updating the Immunisation Equity Strategy to support local services in addressing inequalities in vaccine uptake.

UKHSA will be publishing further briefings, with this report acting as a starting point for future work on health inequalities for health protection. This will include further analysis into more granular geographies, hazard specific inequalities, and the contributing factors to inequalities. It will catalyse action across the health system to improve health outcomes in health security and support all our communities live longer and in better health.

Supplementary data

Download the supplementary data file

References

1. Petersen J and others. ‘Ethnic inequalities in hospital admissions in England: an observational study’ BMC Public Health 2021
2. UKHSA. Immunisations and Vaccine-Preventable Diseases. ‘National Influenza and COVID-19 surveillance report: Week 3 report (up to week 2, 2024 data)’ 2024
3. UKHSA, Immunisations and Vaccine-Preventable Diseases. ‘Seasonal influenza vaccine uptake in GP patients in England: winter season 2023 to 2024’ 2024
4. UKHSA. ‘Surveillance of influenza and other seasonal respiratory viruses in the UK, winter 2023 to 2024’
5. Department of Health and Social Care. Fingertips
6. UKHSA. ‘Heat mortality monitoring report: 2023’
7. UKHSA. ‘HECC 2023 report. Chapter 4: Impacts of climate change and policy on air pollution and human health’
8. National Institutes of Health. ‘Impact of Indoor Air Pollutants on the Cardiovascular Health Outcomes of Older Adults: Systematic Review – PMC’
9. ‘Chief Medical Officer’s Annual Report 2022: Air Pollution’
10. Rizmie D and others. ‘Impact of extreme temperatures on emergency hospital admissions by age and socio-economic deprivation in England’
11. Murage P and others. ‘What individual and neighbourhood -level factors increase the risk of heat-related mortality? A case-crossover study of over 185,000 deaths in London using high-resolution climate datasets’
12. Gasparrinni A and others. ‘Small-area assessment of temperature-related mortality risks in England and Wales: a case time series analysis’
13. Corcuera Hotz I and others. ‘The Effects of Temperature on Accident and Emergency Department Attendances in London: A Time-Series Regression Analysis’
14. Chief Scientist’s Group. ‘Social deprivation and the likelihood of flooding’
15. Public Health England (now UKHSA). ‘The English National Study of Flooding and Health: summary of the evidence generated to date’
16. ‘Fatal Unintentional Non-Fire Related Carbon Monoxide Poisoning: Data from Narrative Verdicts in England and Wales, 1998 to 2019’
17. Close R and others. ‘As safe as houses; the risk of childhood lead exposure from housing in England and implications for public health’ BMC Public Health
18. UKHSA. ‘Impacts on health of emissions from landfill sites’
19. Mueller N and others. ‘Socioeconomic inequalities in urban and transport planning related exposures and mortality: A health impact assessment study for Bradford, UK’ ScienceDirect
20. Race Equality Foundation. ‘How will the climate and nature crises impact people from Black Asian and Ethnic Minority Communities’
21. UKHSA. ‘Adverse weather and health plan equity review and impact assessment 2024’
22. The Health Foundation. ‘Health inequalities in 2040: current and projected patterns of illness by deprivation in England 2024’
23. Razieh C and others. ‘Differences in the risk of cardiovascular disease across ethnic groups: UK Biobank observational study, 2022’
24. Goff LM. ‘Ethnicity and Type 2 Diabetes in the UK, 2019’
25. Aldridge R and others. ‘Household overcrowding and risk of SARS-CoV-2: analysis of the Virus Watch prospective community cohort study in England and Wales’ PMC
26. DHSC. Fingertips
27. Imperial College London. ‘London Health Burden of Current Air Pollution and Future Health Benefits of Mayoral Air Quality Policies’
28. Jephcote C and others. ‘Spatial assessment of the attributable burden of disease due to transportation noise in England, ScienceDirect’
29. UKHSA. ‘Tackling TB in inclusion health groups: a toolkit for a multi-agency approach’
30. UKHSA. ‘Hepatitis C in England and the UK’
31. UKHSA. ‘Shooting Up: infections and other injecting-related harms among people who inject drugs in the UK, data to end of 2021’
32. Public Health England (now UKHSA). ‘Evaluation of hepatitis C test and treat interventions targeted at homeless populations (outside London) in England during the COVID-19 pandemic report 2020’
33. UKHSA. ‘Diphtheria in England: 2022’
34. Bloch A and others. ‘Environmental racism, segregation and discrimination: Gypsy and Traveller sites in Great Britain 2025’
35. Aldridge R and others. ‘Morbidity and mortality in homeless individuals, prisoners, sex workers, and individuals with substance use disorders in high-income countries: a systematic review and meta-analysis’
36. ‘Overall costs of health inequalities. Submission to the Marmot Review’
37. UKHSA. ‘Supporting vulnerable people before and during hot weather: people experiencing homelessness and sleeping rough’

Further information

Methodology and definitions

Infectious disease and infection admissions

Hospital admissions for infectious disease and infections were identified using the Secondary Uses Services Admitted Patient Care (SUS APC) for the year 1 September 2023 to 31 August 2024. They were defined as hospital admissions due to infectious disease or infection based on the primary or secondary diagnosis (ICD-10 codes in 1st or 2nd position only). Admissions were filtered to one episode per patient. We use emergency hospital admissions. This excludes elective and admissions for maternity. The analysis focuses on infectious diseases and infections and includes some non-communicable diseases that occur due to infection, where there is published evidence that infections are its primary aetiological factor (PAF), that is, PAF is over 60%.

Ethnic group

Individual ethnic group data is based on records from the General Practice Extraction Service (GPES) Data for Pandemic Planning and Research (GDPPR), supplemented by ethnicity data recorded in the Hospital Episode Statistics (HES) data sets. We use the NHS Digital HES ethnicity method for assigning ethnic group. We use the most detailed level of ethnic group data available in HES, which includes the 16 ethnic groups, under 5 main categories, in line with the NHS Data Dictionary for ethnic groups. The White British ethnic group is used as the reference group for comparisons because this group represents the largest population in England. The largest group is used because it has the most stable rate and therefore the comparison will be subject to less fluctuation. Population sizes per ethnic group were from ONS Census 2021 data.

Deprivation 

The Index of Multiple Deprivation (IMD) is the official measure of relative deprivation in England produced by the Ministry of Housing, Communities and Local Government (MHCLG). It describes the relative level of deprivation in a small area (lower layer super output area (LSOA)), but not necessarily the individual people living in the area. Many non-deprived people live in deprived areas and many deprived people live in non-deprived areas. Areas are ordered by IMD score and grouped into 5 categories (IMD quintiles). IMD 1 refers to the most deprived areas and IMD 5 refers to the least deprived areas. Deprivation level was assigned to each hospital admission based on the area where the patient lived, using their home postcode at the date of admission. The least deprived group (IMD 5) is used as the reference group throughout.

Age and sex

Individual age is based on age at time of hospital admission, using date of birth. We use 8 age bands:

• 0 to 14 years
• 15 to 24 years
• 25 to 34 years
• 35 to 44 years
• 45 to 54 years
• 55 to 64 years
• 65 to 74 years
• over 75 years

Sex is recorded as either male or female. Age and sex are used to calculate standardised rates of admission, using the European Standard Population.

Region 

An individual’s region is based on their postcode, from APC. Regions are categorised using the 7 England NHS regions.

Admission rate ratios and rate differences

We calculated age- and sex-standardised admission rates, rate ratios and rate differences for deprivation quintiles, ethnic groups and regions. Admission rates were calculated per 100,000 of the population.

Rate ratios were considered disproportionate if they (and +/- 95% confidence intervals on the estimate) are over 1.25 or under 0.80. This follows the Race Disparity Unit’s guidance on disproportionality. 

Admissions attributable to inequality are calculated using the following formulae:

• attributable fraction is equal to the admission rate difference divided by the admission rate in inequality group

• attributable events are equal to the attributable fraction multiplied by the number of admissions in inequality group

Costs to the NHS

Admissions data are from Hospital Episode Statistics for the financial year 6 April 2022 to 5 April 2023. The National Tariff Payment System is used to estimate the cost of each admission based on healthcare resourcing groups (HRG), adjusting for mode of admission, short stay emergency payment, and excess bed days. For admissions not matching to a tariff, the admission is matched to the National Cost Collection using the HRG currency and mode of admission. If the admission is not matched to either the National Tariff Payment System or the National Cost Collection, a cost estimate for the admission is generated by multiplying the length of stay by the average cost per day for a general ward.

CORE20PLUS Framework

The CORE20PLUS framework, developed by NHS England (NHSE), includes those groups we know experience health inequalities across health protection outcomes. Health protection inequalities means that there are differences between communities and population groups, such as risk of exposure to external health hazards, susceptibility to poor outcomes when exposed, and access to and acceptability of health protection interventions.

CORE20 is defined as the most deprived 20% of the population.

PLUS (when added to CORE20) refers to groups known to experience health protection outcome inequalities, determined based on knowledge of the local area, health hazard and how vulnerabilities may interact with each other. These include:

• people experiencing poverty who are not living in areas classified as the most deprived 20%
• some people with protected characteristics as defined by the Equality Act 2010
• people with long term health conditions that place them at increased risk of infection or increased vulnerability to external health hazards
• people providing and receiving social care

Inclusion health groups

Inclusion health is an approach to addressing extreme health inequalities in people and communities who are socially excluded. These groups typically experience very poor health. This includes high risk of infection, stigma and discrimination, poor access to and experience of healthcare and other services. For example:

• people experiencing homelessness
• people with drug and alcohol dependence
• asylum seekers
• refugees and undocumented migrants
• Gypsy, Roma and Traveller communities
• sex workers
• people in contact with the justice system
• victims of modern slavery
• other socially excluded groups

ICD-10 codes and categories

A list of ICD-10 codes for all infections and infectious diseases was created by UKHSA through consolidated literature search, manual code review, and validation by subject matter experts. The list of ICD-10 codes includes codes that are sequalae of infectious diseases or infections in at least 60% of cases, giving a more comprehensive view of inequalities from infectious diseases and infections. Inclusion criteria for the list of ICD-10 codes is presented in the description below:

Inclusion covers:

• an infection is caused by any pathogen entering the body – this includes bacteria, fungi, protozoa, worms, viruses and prions.
• the condition is an infection
• the condition is not an infection but there is published evidence that infections are its primary aetiological factor (that is, the primary aetiological factor (PAF) over 60%)

Exclusion means that the condition:

• is sometimes caused by an infection, but infections are not the primary aetiological factor (that is, the PAF is over 60%)
• can result in an infection but is neither an infection nor caused by one (for example, diabetes)

Disease categories are based on primary mode of transmission, with sorting of individual ICD-10 codes informed by ICD-10 chapters, the World Health Organisation category of communicable diseases, and subject matter expert and clinician input at ICD-10 code level. Categorisation enables analysis of admission rates across different types of disease, to see how inequalities differ across modes of transmission, including:

• respiratory – infections and infectious diseases with respiratory transmission, including influenza, COVID-19 and tuberculosis
• gastrointestinal – infections and infectious diseases with GI transmission, for instance norovirus
• touch and contact – infections and infectious diseases primarily spread through touch or close contact, including mpox and streptococci and staphylococci infections
• sexually transmitted infections and bloodborne viruses – all infections with primarily sexually transmitted pathogens, hepatitis B and C, HIV, and resulting complications
• invasive infections – infections in a sterile site that do not fall into one of the above categories, such as some forms of sepsis and lymphadenitis
• other communicable diseases – communicable diseases that do not fall into one of the above categories, such as vector-borne
• other infections – infections within UKHSA’s remit that don’t fall into one of the above categories, for instance surgical site infections

Not all diseases listed in the ICD-10 code list are present in the admissions data. The ICD-10 code list for the specific diseases is included in the supplementary data tables. Note that the gastroenteritis specific disease category, used in Figure 2 and Table 1, refers to infectious cause gastroenteritis.

Caveats

Many conditions have multiple modes of transmission, the groupings are based on primary mode of transmission. Some conditions are considered vaccine preventable diseases, these are distributed between the categories depending on their primary mode of transmission.

Limitations

ICD-10 code groupings recorded in hospital data for many infections have imperfect sensitivity and specificity, which is likely to lead to some cases being missed and other cases identified as having the infection when they do not. The extent of this inaccuracy will vary by pathogen group. In some instances this misclassification may differ between ethnic groups.  For example, different skin tones may affect interpretation of dermatological manifestations of infections. The analyses should therefore not be relied on to give an accurate picture of the total number of hospitalisations for specific infections.

Positive and Negative predictive values of ICD code lists can also vary over time and across patient groups. In time-periods when infections are very rare, as is the case for some vaccine preventable disease, the Positive predictive value will be lower and the Negative predictive value will be higher than at time periods when the disease is more common. Similarly, in patient groups with higher prevalence of disease, the Positive predictive value will be higher and the Negative predictive value lower than in patients with a lower prevalence of disease. This may make comparison of trends over time or between patient groups, that are based on ICD codes without laboratory confirmation or clinical information, difficult to interpret.

ICD-10 code lists developed for patient populations in other countries have not been validated in this patient population and therefore, published sensitivity, specificity, positive and negative predictive values may not be reliable for this population.

It is important to note that the majority of those eligible are vaccinated, across all groups, but there are variations in coverage by social deprivation and ethnicity. Small differences in uptake can lead to important differences in impact of vaccine preventable infections. Coverage can be affected by multiple barriers across the vaccination pathway including, differential access to healthcare, language barriers, inability to take time off work to attend appointments, lack of access to translated materials, costs of travel to appointments, lack of trust in vaccines or those promoting vaccines, misinformation and different perceptions of risk. 

There may also be different thresholds for hospital admission in different groups, for example children may be more likely to be admitted on a precautionary basis than adults so hospitalisation is not always a good marker of severity. Nevertheless, the analyses show marked differences in rates of hospitalisation for different infections that are of a scale that is unlikely to be caused by these limitations. Using consistent methodology across different infection groupings allows some comparison of which infections show the most marked inequalities.  However we should not be over-reliant on the precise point estimates for rate ratios in specific groups, rather they give an indication of the potential magnitude of inequalities.

There are multiple causes for these variations which merit further investigation so that actions can be targeted to reduce inequalities. Further analyses that link laboratory data and hospital data are merited to explore these issues for specific pathogens and several groups at UKHSA are currently developing these approaches.

Acknowledgements

We are grateful to NHS England, the Office for Health Improvement and Disparities and the Office for National Statistics for provision of the data used in this report.

Authors and contributors

• Dr Meg Scott – Principal Health Analyst
• Luke Prince – Senior Health Analyst
• Sarah Whittle – Health Analyst
• Matt Wells – Senior Health Analyst
• Zoe Richardson – Senior Health Analyst
• Dr Felicity Southworth – Senior Health Analyst
• Dr Tirion Roberts – Senior Health Analyst
• Professor Andrew Hayward – National Lead, Inclusion Health
• Dr Noor Saeed – Head of Health Equity Intelligence
• Amy Jackson – Head of Health Equity Strategy and Engagement
• Monica McCluskey – Head of Health Equity Policy
• Dr Leonora Weil – Deputy Director Health Equity and Inclusion Health
• Dr Shona Arora – Director of Health Equity and Clinical Governance
• UKHSA Analysis and Intelligence Assessment  
• UKHSA Health Equity and Inclusion Health
• UKHSA Strategy and Policy
• UKHSA Chief Medical Advisor Group