Quantifying contributing factors for hospitalization inequalities: influenza and COVID-19, England, 2023 to 2024

Quantifying contributing factors for hospitalization inequalities: influenza and COVID-19, England, 2023 to 2024 | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Quantifying contributing factors for hospitalization inequalities: influenza and COVID-19, England, 2023 to 2024 Meg Scott , Lucy Browning , Tirion Roberts , Zoe A. Richardson doi: https://doi.org/10.1101/2025.08.14.25333430 Meg Scott 1 Chief Data Officer Group, UK Health Security Agency , UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: megan.scott{at}ukhsa.gov.uk Lucy Browning 1 Chief Data Officer Group, UK Health Security Agency , UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Tirion Roberts 1 Chief Data Officer Group, UK Health Security Agency , UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zoe A. Richardson 1 Chief Data Officer Group, UK Health Security Agency , UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF Abstract Health system decision makers need information on the relative contribution of modifiable contributing factors to inequalities, to develop and prioritise interventions. We used granular linked administrative data to examine ethnic group and deprivation inequalities in hospitalizations for COVID-19 and influenza, England, 2023 to 2024. We used logistic regression to quantify the overall contribution to inequality of multiple contributing factors and estimated the relative contribution of factors which can be influenced by policy: vaccination uptake, pre-existing health conditions and deprivation. Most, but not all, of the excess odds of hospitalization were due to differences between groups in contributing factors considered. For ethnic groups, accounting for vaccination and deprivation made the biggest difference to the excess odds of hospitalization, more so than accounting for a pre-existing health condition. For deprivation, accounting for vaccination made the biggest difference to the excess odds of hospitalization, followed by a pre-existing health condition. Reducing inequalities in vaccination uptake and pre-existing health conditions would reduce hospitalization inequalities; reducing vaccination differences may reduce inequalities the most. For ethnic group inequalities, reducing deprivation inequalities is also vital to reduce hospitalization inequalities. Introduction Health policy can be targeted to reduce health inequalities. To do this effectively, health system decision makers need current, consistent and comparable metrics on inequalities, to provide a baseline. They also need information on the relative importance of contributing factors to inequalities, especially factors which are, to some extent, modifiable by health policy. This can be challenging without sufficiently detailed data - improving data on different ethnic groups and deprivation areas is a key priority for health protection in England [ 1 ]. Influenza and COVID-19 are significant contributors to morbidity and mortality in England, and negatively impact the national health system annually [ 2 ]. The burden is far greater for certain groups, for both influenza and COVID-19 infection, and for severe outcomes following infection, including hospitalization, intensive care admission and mortality [ 3 - 6 ]. Influenza and COVID-19 have established seasonal vaccination programmes, with wide coverage and notable inequalities in uptake [ 7 , 8 ]. However, inequalities are not systematically tracked in detail for these diseases. There is limited current evidence on the relative importance of contributing factors, such as pre-existing health conditions or vaccination uptake, in accounting for group differences. To date, evidence on inequalities in health outcomes for influenza and COVID-19 varies by time period covered, health outcomes and dimensions of inequalities observed. Inequalities are known for hospitalization, vaccination uptake and pre-existing health conditions but these are often monitored separately, rather than linking data across the same individuals. Previous UK Health Security Agency analysis [ 9 ] confirmed significant differences in hospitalization rates for COVID-19 and influenza between ethnic groups and deprivation areas (England, winter 2022 to 2023). Influenza admission rates for the most deprived areas were 2.6 times higher than the least deprived areas, and 2.1 times higher for COVID-19. For influenza (but not COVID-19) there were persistent differences in hospitalization rates between ethnic groups – the Pakistani group had admission rates for influenza 2.7 times higher than the White group. Post-pandemic, ethnic group inequalities for COVID-19 have been comparably lower [ 10 ]. Vaccination protects against the risk and severity of infection, reducing the risk of hospitalization. Vaccine uptake is consistently lowest in the most deprived areas as compared to the least deprived areas. For the mid-winter season, 2023 to 2024, for COVID-19, 79.3% of those aged 65 years or over from the least deprived areas were vaccinated in the 6 months prior, compared to 59.1% in the most deprived areas [ 7 ]. In 2023 to 2024, influenza vaccine uptake was also lower in the most deprived areas compared to the least deprived areas, across all vaccination cohorts [ 8 ]. There were significant ethnic group inequalities; for COVID-19, vaccine uptake was lowest for the Pakistani (18.1%), and Bangladeshi groups (24.4%), and highest for the White British (75.5%) and White Irish groups (64.1%) [ 7 ]. For influenza, vaccine uptake was lowest for the Black or Black British Caribbean (45.5%) and the Black or Black British African groups (48.0%), and highest for the White British (81.7%) and White Irish groups (75.9%) [ 7 ]. Pre-existing health conditions which predispose individuals to a higher risk from COVID-19 and influenza are detailed in the Clinical risk groups in the Green Book, which provides information for public health professionals on immunization [ 11 ]. Risks for many of these conditions are higher among certain ethnic groups, including cardiovascular disease, diabetes and multimorbidity [ 12 - 14 ]. People from Indian, Pakistani, Bangladeshi, Black African, Black Caribbean and people belonging to “Any other Black ethnic group”, “Any other Asian ethnic group”, or mixed ethnic groups may also have higher risk of developing multiple long-term conditions [ 15 ]. Pre-existing health conditions are also highest for those living in deprived areas, with large differences by deprivation for mortality involving lung cancer and asthma, two risk factors for COVID-19 and influenza [ 12 ]. Some existing evidence shows that reducing differences in contributing factors across groups would reduce inequalities in health outcomes. However, it is challenging to untangle which contributing factors account for the most inequality. Analysis on COVID-19 mortality during the pandemic demonstrated that a large proportion of the difference in risk between ethnic groups was explained by demographic, geographic and socioeconomic factors [ 16 ]. However, significant differences remained. Trends in COVID-19 inequalities have changed significantly since the pandemic [ 17 , 18 ] and more recent time periods are less well studied. There is limited evidence for influenza outcomes, or the role played by other contributing factors. We used detailed administrative data sources on hospitalization, vaccination and laboratory testing, linked at individual patient-level. We estimated relative odds of hospitalization for COVID-19 and influenza, in the time period 1 September 2023 to 31 March 2024, for different ethnic groups and deprivation areas. We set out repeatable methods for quantifying inequalities and used logistic regression to quantify the overall contribution to inequality of a range of contributing factors and then estimated the relative contribution of factors which can be influenced by policy: vaccination uptake, pre-existing health conditions and deprivation. We focused on inequalities in hospital admissions as an indicator of more severe disease, and a common metric for burden on the health system in England. Methods A range of data sources were linked. This retrospective cohort study includes individuals in England who were eligible for vaccination for COVID-19 or influenza, between 1 September 2023 and 31 March 2024. Data sources Hospitalization data was from the Secondary-Use-Services (SUS) Admitted Patient Care (APC) dataset [ 19 ] from National Health Service (NHS) England, used to identify individuals admitted to hospital between 1 September 2023 and 31 March 2024. Vaccination data was from the Immunisation Information System (IIS) registry of all individuals registered with a general practitioner in England. This holds population denominator data for the vaccine eligible cohort in England, and corresponding vaccination events. We looked at all vaccination events between 1 September 2023 and 31 March 2024 – the vaccination roll out for winter 2023 to 2024 began on 1 September 2023. COVID-19 and influenza positive tests were linked to the patient-level data from SUS using the Second Generation Surveillance System (SGSS) from the UK Health Security Agency (UKHSA). This is a national database that captures routine laboratory test results on infectious diseases. Patients were linked across admissions, vaccination and testing data using unique NHS numbers. Data was processed in accordance with NHS Digital Data Sharing Agreements, accessed only on a secure restricted platform by approved and trained researchers. Admission rates were calculated at group level with no individual patients identified. Study population The study population was defined using the IIS registry, with cohorts identified using age and at-risk flags, based on pre-existing health conditions. Eligibility criteria for vaccination is set by the Joint Committee on Vaccination and Immunisation (see supplementary 1). Analysis was restricted to the vaccine eligible cohort to estimate the role of vaccination in contributing to the risk of hospitalization. This means that differences between groups in vaccination uptake and admission are not due to differences in eligibility criteria, but due to whether individuals take up the offer of vaccination. For robustness, we compared inequalities in admission rates for the eligible population against the whole population. Patterns of admission rate ratios were consistent between the eligible cohort and whole population. However, the relative contribution of individual factors may differ between the eligible cohort and the whole population. Conclusions of this paper apply to the vaccine-eligible cohort only. Outcome and covariates Our primary outcome was hospitalization due to influenza or COVID-19, between 1 September 2023 and 31 March 2024. Hospitalizations were identified at patient level, separately for influenza and COVID-19. If an individual was hospitalized multiple times during the period, only the first hospital spell was included. Where patients were treated under multiple consultant episodes within a spell, we used diagnoses data from the first episode, to exclude hospital acquired infections. This included all admissions coded as an emergency were included, we excluded elective admissions, maternity and admissions for birth (admission codes in supplementary 2). Admissions were linked to laboratory testing data from SGSS to validate recorded diagnoses. Due to the size of the population, we aggregated data into counts of admissions. Influenza admissions were defined by a primary ICD-10 diagnosis code for influenza, or any admission with a respiratory primary diagnosis code, if also linked to a positive influenza test up to 14 days prior or 2 days post admission date. COVID-19 admissions were defined by a primary ICD-10 diagnosis code for COVID-19, or any admission with a respiratory primary diagnosis code, if also linked to a positive COVID-19 test up to 2 days prior or 1 day post hospital admission date. (ICD-10 codes in supplementary 3). The different definitions were consistent with the protocol for surveillance of influenza and other seasonal respiratory viruses, UK, winter 2023 to 2024 [ 20 ]. Vaccination was defined as a binary variable (0/1) based on patient vaccination status at the time of hospital admission. This was coded ‘1’ if an individual was vaccinated at least 14 days (influenza) or 7 days (COVID-19) prior to hospital admission, and ‘0’ otherwise, between 1 September 2023 to 31 March 2024. Only the first vaccination was counted if multiple vaccination records existed. For non-hospitalized individuals, vaccination status was determined by any recorded vaccination within the study period. Pre-existing health condition was coded as a binary variable (0/1) based on whether an individual was flagged as ‘low-risk’ (0) or ‘high-risk’ (1) in IIS data. ‘High-risk’ was any individual within a clinical at-risk group. Deprivation area was recorded using quintiles from the Index of Multiple Deprivation (IMD) 2019 [ 21 ]. This was assigned based on patient home postcode at date of admission. Deprivation quintiles were ordered from most to least deprived, IMD1 to IMD 5. IMD 5 (least deprived) was the reference group throughout, coded as a categorical variable. Individual ethnic group categories were 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) datasets, where GPES data was missing. Where multiple ethnicity records existed for an individual, records were selected following OHID methodology [ 22 ]. We used the most detailed level of ethnic group data available in HES, 16 ethnic groups in line with the NHS Data Dictionary [ 23 ].The White British group was used as the reference group throughout because this group has the largest population in England. Individual age was based on age at time of admission, aggregated to five age bands: 0-4 years, 5-14 years, 15-44 years, 45-64 years, 65+ years. Sex was recorded as either male or female, based on IIS, or APC if missing. Region was based on patient postcode, from IIS, or APC if missing. We used the nine statistical regions; North East, North West, Yorkshire and The Humber, East Midlands, West Midlands, East of England, London, South East and South West. Exclusions We excluded records where an admission was not identifiable as either influenza or COVID-19 through either diagnosis code or linked laboratory test, or if the patient record was missing information on any of the covariates used. The number of excluded individuals was 1,430,256 for influenza and 643,336 for COVID-19, representing 0.05% and 0.03% of the study populations, respectively (see supplementary 4). Data analyses We calculated admission rates and rate ratios for each deprivation quintile, relative to the least deprived quintile (IMD 5), and for each ethnic group, relative to the White British group. Admission rates were directly age-and sex-standardized admission rates per 100,000 of the population, for influenza and COVID-19 separately. Denominators were based on the eligible cohort, using IIS data. Rate ratios were considered disproportionate if they (+/-95% Confidence Intervals (CI)) are >1.25 or <0.80. This follows the Race Disparity Unit’s guidance on disproportionality [ 24 ]. We used binomial regression to estimate the proportion of each group hospitalized due to influenza or COVID-19, allowing us to estimate the relative odds for each group: Where y i is the number of admissions of individuals in a sub-group i, n i is the count of individuals in the sub-group and p i is the probability of hospitalisation for i . Where p ( y ) is the probability of being admitted to hospital, b are coefficients of the independent variables, and the x are the independent variables. We used model comparisons to quantify how estimated odds ratios for ethnic groups and deprivation quintiles changed depending on which variables were included in models, following suggested outcome models for epidemiology [ 25 ]. We compared a baseline model (separate for ethnic group and deprivation) to a full model including all contributing factors together. We assessed the relative individual contribution of a subset of contributing factors: vaccination uptake, pre-existing health condition and deprivation. These were selected as the contributing factors which could be influenced by policy. Influenza and COVID-19 are modelled separately. Inequalities between ethnic groups are compared separately to inequalities between deprivation quintiles: ‐ Model 1a (baseline: ethnic group inequalities): age, sex, ethnic group ‐ Model 1b (baseline: deprivation inequalities): age, sex, deprivation (IMD) ‐ Model 2 (full model): age, sex, ethnic group, vaccination uptake, pre-existing health condition, deprivation, region ‐ Model 3: full model minus vaccination uptake ‐ Model 4: full model minus deprivation ‐ Model 5: full model minus pre-existing health condition ‐ Model 6: full model minus ethnic group* * Ethnic group was considered as an individual contributing factor, for comparison to the baseline model for deprivation inequalities. This was for comparison only, ethnic group was not considered a modifiable contributing factor. The models were fit using glm with binomial regression, from the stats package in R[ 26 ]. Model predictions were extracted using the marginaleffects R package [ 27 ], and odds ratios from the modelsummary package [ 28 ]. All analyses were carried out using R version 4.3.2. Differences between odds ratios estimated from different models were considered statistically significant if 95% CI were non-overlapping. Odds ratios were considered to indicate disproportionality when they (and 95% CI) were >1.25 or <0.8, following the guidance from the Race Disparity Unit [ 24 ]. Results Descriptive analysis There were 28,898,935 people identified in the eligible cohort for vaccination for influenza, and 18,905,708 people for COVID-19, with 21,840 COVID-19 hospitalizations and 18,917 influenza hospitalizations. There were significant inequalities between ethnic groups and deprivation in hospitalizations, vaccination uptake, pre-existing health condition ( Table 1 and 2 for study cohort, exclusions in supplementary 4). We restrict observations to one per individual. View this table: View inline View popup Download powerpoint Table 1. Descriptive metrics for the vaccine-eligible cohort: COVID-19 View this table: View inline View popup Download powerpoint Table 2. Descriptive metrics for the vaccine-eligible cohort: influenza Age and sex-standardized rates: influenza Four ethnic groups had disproportionately high influenza admission rates compared to the White British group ( Fig. 1a ). The Pakistani group had the highest rate, 2.5 (95% CI: 2.32, 2.75) times higher than the White British group, equating to 76 (95% CI: 65.93, 86.86) more admissions per 100,000 population. Only the Chinese group had disproportionately lower rates, 0.5 times (95%CI: 0.35, 0.73) the White British group, equating to 25 (95% CI: 15.21, 33.94) fewer admissions per 100,000. Download figure Open in new tab Figure 1. Admission rate ratios by ethnic group, influenza and COVID-19, vaccine- eligible population, England, winter 2023 to 2024 People living in more deprived areas had a disproportionately high influenza admission rate compared to people living in less deprived areas ( Fig. 2a ). The admission rate for the most deprived areas was 1.9 (95% CI: 1.75, 2.02) times higher than the least deprived, equating to 36 (95% CI: 32.05, 39.38) more admissions per 100,000. Download figure Open in new tab Figure 2. Admission rate ratios by deprivation level, influenza and COVID-19, vaccine- eligible population, England, winter 2023 to 2024 Age and sex-standardized rates: COVID-19 No ethnic minority groups had disproportionately high COVID-19 admission rates compared to the White British group ( Fig. 1b ). Only the ‘Any other Black/African/Caribbean background’ group had disproportionately low rates of 0.6 (95% CI: 0.46, 0.79) times the White British group, equating to 36 (95% CI: 21.18, 50.66) fewer admissions per 100,000. People living in more deprived areas had higher COVID-19 admission rates, however the trend was not disproportionate ( Fig. 2b ). Logistic regression: influenza Ethnic groups The odds ratios (ORs) for ethnic groups are in Table 3 . Odds ratios for all variables in Supplementary 8. View this table: View inline View popup Download powerpoint Table 3. Odds ratios from full and baseline model: Influenza, ethnic group focus For influenza, for ethnic group inequalities, in the full model only the Pakistani group maintained disproportionately high odds of influenza hospitalization. Controlling for vaccination uptake, pre-existing health conditions, deprivation and region together resulted in significantly smaller estimated odds of influenza hospital admission in four of 15 ethnic minority groups; Pakistani, Black Caribbean, Bangladeshi, and Black African, as compared to baseline odds ratios ( Fig. 3a ). The largest reduction was for the Pakistani group where the estimated odds ratio decreased from 2.2 (95% CI: 2.07, 2.36) to 1.6 (95% CI: 1.48, 1.70). Download figure Open in new tab Figure 3. Odds ratios for full model versus baseline model, influenza and COVID-19, ethnic groups, vaccine-eligible population, England, winter 2023 to 2024 The overall event prevalence of influenza hospitalization was 0.06%, meaning approximately 6 hospitalizations occur per 10,000 individuals, within the study period. Because hospitalization with influenza is a rare outcome, the odds ratios approximate risk ratios. Changes in excess odds can also be interpreted as meaningful proxies for excess risk. For example, for the Pakistani group, 50% of the excess odds were associated with the contributing factors considered together: vaccination uptake, pre- existing health condition, deprivation and region factors ( Table 3 ). Considering the subset of contributing factors individually, adjusting for vaccination uptake and deprivation level both reduced the estimated odds ratio of influenza hospital admission for ethnic groups, more than adjusting for pre-existing health conditions (Ethnic group ORs: Table 5 , all ORs: supplementary 10). Interpreted as the change in excess odds, we compared the excess odds for the full model, against the excess odds minus the factor, against the baseline odds. For the Pakistani group for the full model comparison, up to 14% of the excess odds were likely to be associated with vaccination uptake, up to 13% with deprivation and up to 3% with pre-existing health conditions. Deprivation areas For deprivation, even after adjusting for contributing factors, the most deprived areas still had disproportionately high odds of admission for influenza. The full model estimated significantly smaller odds ratios for more deprived areas compared to the least deprived areas, as compared to the baseline model ( Fig. 4a ). The largest reduction was in the most deprived group where the estimated odds ratio decreased by from 1.9 (95% CI: 1.83, 2.00) to 1.5 (95% CI: 1.46, 1.60). Expressed in risk ratios, accounting for contributing factors of vaccination uptake, pre-existing health condition, region and ethnic group reduced the excess odds by 44% (Deprivation ORs: Table 4a , all ORs: supplementary 9). View this table: View inline View popup Download powerpoint Table 4a. Odds ratios from full and baseline model: Influenza, deprivation focus. Download figure Open in new tab Figure 4. Odds ratios for full model versus baseline model, influenza and COVID-19, deprivation levels, vaccine-eligible population, England, winter 2023 to 2024 Considering risk factors individually, adjusting for vaccination status and pre-existing health conditions reduced the estimated odds ratio of influenza hospital admissions for more deprived areas, more than adjusting for ethnic group (Deprivation ORs: Table 5 , all ORs: supplementary 10). For influenza, for the most deprived group, up to 17% of the excess odds compared to the least deprived group were likely to be associated with vaccination status, 16% was associated with pre-existing health condition, and 7% with ethnic group. Logistic regression: COVID-19 The overall event prevalence of COVID-19 hospitalization was 0.11%, meaning approximately 11 hospitalizations occur per 10,000 individuals. Because hospitalization with COVID-19 is a rare outcome, the odds ratios approximate risk ratios. Changes in excess odds can also be interpreted as meaningful proxies for excess risk. Ethnic groups Controlling for vaccination uptake, pre-existing health conditions, deprivation, and region together, did significantly reduce the estimated odds ratio of COVID-19 hospital admission for eight of 15 ethnic groups ( Fig. 3b ), but odds ratios were not disproportionately high for any ethnic group, in either the baseline or full model. Therefore, we do not present the excess odds here (All ORs: Supplementary 8). Considering risk factors individually, adjusting for vaccination status reduced the estimated odds ratio of influenza hospitalizations for ethnic groups, more than adjusting for deprivation area or pre-existing health conditions (Fig. 5b). Odds ratios were not disproportionately high as compared to the White British group, for any of the models specified (ORs: Supplementary 11). Deprivation areas After adjusting for all risk factors together, the estimated odds of hospitalization for COVID-19 for more deprived areas were no longer disproportionately high compared to the least deprived areas ( Fig. 4b ). The odds ratio for the most deprived quintile reduced from 1.4 (95% CI: 1.38, 1.50) to 1.0 (95% CI: 0.99-1.08). In terms of excess odds, this was fully accounted by the contributing factors of vaccination uptake, pre-existing health condition, ethnic group and region (Deprivation ORs: Table 4b , all ORs: Supplementary 9). View this table: View inline View popup Download powerpoint Table 4b. Odds ratios from full and baseline model: COVID-19, deprivation focus Considering the subset of contributing factors individually, adjusting for vaccination uptake reduced the estimated odds ratio of hospitalization for more deprived areas more than adjusting for pre-existing health conditions. Both led to a greater reduction than adjusting for ethnic group (Deprivation ORs: Table 6 , all ORs: Supplementary 11). For the most deprived group, expressed in excess odds, up to 78% of the excess odds were likely to be associated with vaccination uptake, 48% was likely to be associated with pre-existing health conditions, and accounting for ethnic group increased the higher risk of hospitalization. View this table: View inline View popup Download powerpoint Table 5. Odds ratios from full model minus factor of interest: Influenza View this table: View inline View popup Download powerpoint Table 6. Odds ratios from full model minus factor of interest: COVID-19 Model fit Model fit statistics and checks are included in Supplementary 5. The Akaike’s Information Criterion (AIC) and Bayesian Information Criterion (BIC) significantly improved between the baseline models and the full model, for both influenza and COVID-19. For the full model, which had the best fit, variance inflation factors (VIFs) for all covariates were below 5. This indicates that multicollinearity was not a concern according to standard thresholds [ 29 , 30 ] and the estimated coefficients are unlikely to be unstable due to high correlations among predictors. Discussion Within the vaccine-eligible cohort for influenza and COVID-19, inequalities in hospitalization for influenza were greater than for COVID-19. Comparing directly age- and sex-standardized admission rates, people from the Pakistani and Bangladeshi groups were more than twice as likely to be hospitalized for influenza compared to the White British group. By contrast, COVID-19 age- and sex- standardized admission rates for most ethnic groups were comparable to the White British group, with the “Any other Black/African/Caribbean ethnic group” having disproportionately lower admission rates. People from more deprived areas were more likely to be hospitalized for both influenza and COVID-19, compared to the least deprived, but differences were bigger for influenza. Adjusting for contributing factors significantly reduced differences in the estimated odds of hospitalization for both influenza and COVID-19. For ethnic groups, vaccination uptake and deprivation were associated with the greatest reduction in the estimated odds of hospitalization. For deprivation, vaccination uptake was associated with the greatest reduction. It is likely that a significant proportion of inequality in hospitalization is attributable to group variation in vaccination status. Vaccination uptake is an ideal candidate for public health interventions, as a modifiable risk factor that would reduce inequalities for influenza and COVID-19. Patterns for differences in pre-existing health conditions were more variable across diseases and groups. Pre-existing health condition differences do appear to be a key driver for deprivation differences in influenza admissions - adjusting for pre-existing health conditions reduced the excess odds by up to 16%. Pre-existing health conditions can be a modifiable risk factor, but reducing their prevalence is likely to be more challenging, requiring wider systemic change. For influenza, differences in deprivation area between ethnic groups had a similar effect size to vaccination uptake, reducing the excess odds by up to 30% ( Table 5 ). Reducing absolute levels of poverty, and inequalities in deprivation is likely to have widespread benefits but is likely to be significantly more challenging to change, as compared to increasing vaccination uptake. We found minimal evidence of inequalities for ethnic groups for COVID-19 hospital admissions. This is in stark contrast to early stages of the pandemic (2020 - 2021) when most ethnic groups had significantly worse outcomes from infection by SARS-CoV-2, compared to the White British group. Given widespread evidence of current ongoing ethnic group inequalities in respiratory disease [ 31 ], this is perhaps surprising. However, this finding is congruent with other research; the ONS found no evidence of ethnic minority groups having higher COVID-19 mortality compared to the White British group during the Omicron period (Jan - Dec 2022) [ 17 ]. Historically, increased risk of infection was the main driver of disproportionate clinical outcomes for COVID-19 in ethnic minority groups, with smaller differences observed when considering subsequent outcomes, including hospitalization [ 6 ]. Ethnic minority groups may have higher existing immunity to SARS-CoV-2 because of the higher likelihood of previous infection [ 32 ]. Influenza may show more pronounced ethnic group inequalities compared to COVID-19 because there is sufficient time between infection peaks for immunity to wane, reducing the protective effect on ethnic minority groups. The role of infection in inequalities emphasizes the importance of vaccine programme interventions to maintain immunity. This study provides unique insights into inequalities. The use of detailed administrative data allowed us to create individual patient profiles with rich information on ethnic group and deprivation. Ethnic group data is at a more detailed level than in many surveillance sources, which avoids obscuring trends by aggregating ethnic groups. Linkage of data across sources for vaccination, health history and hospitalization allowed more detail on the interaction between risk factors and hospitalization. However, the study has several limitations. Firstly, our conclusions are limited to those eligible for vaccination. Second, we did not examine inequalities in the risk of infection, access to, or experience of, healthcare services. Our conclusions are limited to people who were hospitalized with influenza or COVID-19 and cannot quantify the role of differences in exposure to illness, severity of illness, access to healthcare, or unequal treatment, for instance, following racism or discrimination [ 33 ]. We also do not consider outcome after hospitalization, for instance, length of stay, admission to critical care, or death. These results represent a snapshot in time in England, so may not be generalisable to other countries, or other time periods. Further work is needed to understand and reduce known inequalities in the wider determinants of health, including in occupational exposure to infection [ 34 - 36 ], poor housing and neighbourhood conditions, including air pollution [37, 38]. People from ethnic minority groups and more deprived areas are likely to experience multiple and cumulative effects of inequalities. Other dimensions of inequality are equally important, such as the disproportionate impact on inclusion health groups. Driving vaccination uptake in groups with low uptake and reducing the prevalence of pre-existing health conditions would reduce hospitalization inequalities. Reducing differences in vaccination uptake may reduce inequalities the most. Appropriate vaccine communication and implementation for all groups should be a policy priority. For ethnic group inequalities, reducing absolute poverty and inequalities in deprivation would also reduce hospitalization inequalities. This is more challenging for policy but remains crucial. Deprivation inequalities have demonstrable consequences for health outcomes and contribute to complex and multiple disadvantages over the life course. Data availability statement The UK Health Security Agency (UKHSA) operates a robust governance process for applying to access protected data that considers: the benefits and risks of how the data will be used compliance with policy, regulatory and ethical obligations data minimization how the confidentiality, integrity, and availability will be maintained retention, archival, and disposal requirements best practice for protecting data, including the application of ‘privacy by design and by default’, emerging privacy conserving technologies and contractual controls Access to protected data is always strictly controlled using legally binding data sharing contracts. UKHSA welcomes data applications from organizations looking to use protected data for public health purposes. To request an application pack or discuss a request for UKHSA data you would like to submit, contact DataAccess{at}ukhsa.gov.uk . Financial support The authors were employed by the UKHSA but received no specific funding for this study. Conflict of interest The authors declare none. Acknowledgements We would like to thank NHS Digital for provision of the data used in this report. We would like to thank reviewers and contributors from the following UKHSA teams: Public Health Analysis, Health Equity and Inclusion Health, Statistics, Infectious Disease Modelling, Immunisations and Vaccine Preventable Diseases. References 1. ↵ UK Health Security Agency , Health equity for health security strategy 2023 to 2026. 2024, UK Health Security Agency : https://www.gov.uk/government/publications/ukhsa-health-equity-for-health-security-strategy-2023-to-2026 . 2. ↵ UK Health Security Agency , Infectious diseases impacting England: 2025 report . 2025 : https://www.gov.uk/government/publications/infectious-diseases-impacting-england-2025-report . 3. ↵ Zhao , H. , et al. , Ethnicity, deprivation and mortality due to 2009 pandemic influenza A(H1N1) in England during the 2009/2010 pandemic and the first post-pandemic season . Epidemiology and Infection , 2015 . 143 ( 16 ). 4. Chandrasekhar , R. , et al. , Social determinants of influenza hospitalization in the United States . Influenza and other respiratory viruses , 2017 . 5. Davidson , J. , et al. , Ethnic differences in the incidence of clinically diagnosed influenza: an England population-based cohort study 2008-2018 . Wellcome Open Research , 2021 . 6. ↵ Irizar , P. , et al. , Ethnic inequalities in COVID-19 infection, hospitalisation, intensive care admission, and death: a global systematic review and meta-analysis of over 200 million study participants . eClinicalMedicine , 2023 . 57 . 7. ↵ UK Health Security Agency , National Influenza and COVID-19 surveillance report: week 3 report (up to week 2 2024 data) . 2024 , UK Health Security Agency, Immunisations and Vaccine-Preventable Diseases . 8. ↵ UK Health Security Agency , Seasonal influenza vaccine uptake in GP patients in England: winter season 2023 to 2024 . 2024 , UK Health Security Agency, Immunisations and Vaccine-Preventable Diseases . 9. ↵ UK Health Security Agency , Inequalities in emergency hospital admission rates for influenza and COVID-19, England: September 2022 to February 2023 . 2023 : https://www.gov.uk/government/publications/covid-19-and-flu-inequalities-in-emergency-hospital-admission-rates . 10. ↵ Public Health England , COVID-19: review of disparities in risks and outcomes . 2020 , PHE publications gateway number: GW-1447 : https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes . 11. ↵ Ramsay , M.E. , Immunisation against infectious disease, U.H.S. Agency, Editor . 2024 : Immunisation against infectious disease . 12. ↵ Bosworth , M. , et al. , Inequalities in mortality involving common physical health conditions, England: 21 March 2021 to 31 January 2023 , H.D. Office for National Statistics, Editor . 2023 : https://www.ons.gov.uk/releases/sociodemographicandgeographicaldifferencesinratesofhospitaladmissionandmortalityforcommonphysicalhealthconditionsengland21march2021to31january2023 . 13. Petersen , J. , J. Kandt , and P.A. Longley , Ethnic inequalities in hospital admissions in England: an observational study . BMC Public Health , 2021 . 14. ↵ The Health Foundation . Inequalities in diagnosed health conditions by ethnicity . 2025 [cited 2025 27/03/2025]. 15. ↵ Hayanga , B. , M. Stafford , and L. Bécares , Ethnic inequalities in multiple long-term health conditions in the United Kingdom: a systematic review and narrative synthesis . BMC Public Health , 2023 . 23 . 16. ↵ Office for National Statistics , Updating ethnic contrasts in deaths involving the coronavirus (COVID-19), England: 8 December 2020 to 1 December 2021 . 2022 . 17. ↵ Office for National Statistics , Updating ethnic and religious contrasts in deaths involving the coronavirus (COVID-19), England: 24 January 2020 to 23 November 2022 . 2023 , Office for National Statistics : https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/articles/updatingethniccontrastsindeathsinvolvingthecoronaviruscovid19englandandwales/24january2020to23november2022 . 18. ↵ Office for National Statistics , Updating ethnic contrasts in deaths involving the coronavirus (COVID-19), England: 10 January 2022 to 16 February 2022 . 2022 , Office for National Statistics : https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/articles/updatingethniccontrastsindeathsinvolvingthecoronaviruscovid19englandandwales/10january2022to16february2022 . 19. ↵ Herbert , A. , et al. , Data resource profile: Hospital episode statistics admitted patient care (HES APC) , I.J.o. Epidemiology, Editor . 2017 . 20. ↵ UK Health Security Agency , Surveillance of influenza and other seasonal respiratory viruses in the UK, winter 2023 to 2024 .. 21. ↵ Ministry of Community Housing and Local Government , The English indices of deprivation . 2019 , Ministry of Community, Housing and Local Government, MCHLG : https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/835115/IoD2019_Statistical_Release.pdf . 22. ↵ Office for Health Improvement and Disparities , Method for assigning ethnic group in the COVID-19 health inequalities monitoring for England (CHIME) tool . 2023 : gov.uk. 23. ↵ NHS Digital , Data quality of protected characteristics and other vulnerable groups: ethnicity . 2022 : https://digital.nhs.uk/data-and-information/data-collections-and-data-sets/data-sets/mental-health-services-data-set/submit-data/data-quality-of-protected-characteristics-and-other-vulnerable-groups/ethnicity . 24. ↵ Race Disparity Unit , Using relative likelihoods to compare ethnic disparities . 2020 : gov.uk . 25. ↵ Greenland S , Daniel R , and P. N, Outcome modelling strategies in epidemiology: traditional methods and basic alternatives . International Journal of Epidemiology , 2016 . 45 . 26. ↵ “The R Stats Package” , R.C.T.a.c. worldwide, Editor . 2024 : https://search.r-project.org/R/refmans/stats/html/00Index.html . 27. ↵ Arel-Bundock , V. , “marginaleffects: Predictions, comparisons, slopes, marginal means, and hypothesis tests,” . 2024 : https://marginaleffects.com/ . 28. ↵ Arel-Bundock , V. , modelsummary: data and model summaries . 2022 : R.J Stat Softw . 29. ↵ Kutner , M. , et al. , Applied linear statistical models . Vol. 5th Edition. 2005 , New York: McGraw-Hill Irwin . 30. ↵ O’brien , R. , A caution regarding rules of thumb for variance inflation factors . Quality and Quantity , 2007 . 41 ( 673-690 ). 31. ↵ UK Health Security Agency , Health inequalities in health protection report . 2025 : https://www.gov.uk/government/publications/health-inequalities-in-health-protection-report . 32. ↵ Pan , D. , et al. , Are clinical outcomes from COVID-19 improving in ethnic minority groups? eClinicalMedicine , 2023 . 61 . 33. ↵ National Health Service England , Improving access for all: reducing inequalities in access to general practice services . 2018 , National Health Service England, NHS England : https://www.england.nhs.uk/wp-content/uploads/2017/07/inequalities-resource-sep-2018.pdf . 34. ↵ Pearce , N. , et al. , Occupational differences in COVID-19 incidence, severity, and mortality in the United Kingdom: Available data and framework for analyses . Wellcome Open Research , 2023 . 35. Rhodes , S. , et al. , Occupational differences in SARS-CoV-2 infection: analysis of the UK ONS COVID-19 infection survey . Journal of epidemiology and community health , 2022 . 76 : p. 841 – 846 . OpenUrl Abstract / FREE Full Text 36. ↵ Balogun , B. , F. Rankl , and W. Wilson , Health inequalities: cold or damp homes . 2023 , Library, House of Commons: Library, House of Commons . https://www.ethnicity-facts-figures.service.gov.uk/housing/housing-conditions/housing-with-damp-problems/latest/ . View the discussion thread. Back to top Previous Next Posted August 21, 2025. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Quantifying contributing factors for hospitalization inequalities: influenza and COVID-19, England, 2023 to 2024 Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Quantifying contributing factors for hospitalization inequalities: influenza and COVID-19, England, 2023 to 2024 Meg Scott , Lucy Browning , Tirion Roberts , Zoe A. Richardson medRxiv 2025.08.14.25333430; doi: https://doi.org/10.1101/2025.08.14.25333430 Share This Article: Copy Citation Tools Quantifying contributing factors for hospitalization inequalities: influenza and COVID-19, England, 2023 to 2024 Meg Scott , Lucy Browning , Tirion Roberts , Zoe A. Richardson medRxiv 2025.08.14.25333430; doi: https://doi.org/10.1101/2025.08.14.25333430 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Epidemiology Subject Areas All Articles Addiction Medicine (569) Allergy and Immunology (863) Anesthesia (300) Cardiovascular Medicine (4442) Dentistry and Oral Medicine (444) Dermatology (383) Emergency Medicine (609) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1511) Epidemiology (15230) Forensic Medicine (30) Gastroenterology (1126) Genetic and Genomic Medicine (6610) Geriatric Medicine (668) Health Economics (998) Health Informatics (4542) Health Policy (1370) Health Systems and Quality Improvement (1613) Hematology (543) HIV/AIDS (1266) Infectious Diseases (except HIV/AIDS) (15923) Intensive Care and Critical Care Medicine (1103) Medical Education (623) Medical Ethics (147) Nephrology (668) Neurology (6607) Nursing (346) Nutrition (999) Obstetrics and Gynecology (1146) Occupational and Environmental Health (957) Oncology (3338) Ophthalmology (974) Orthopedics (369) Otolaryngology (420) Pain Medicine (436) Palliative Medicine (130) Pathology (665) Pediatrics (1693) Pharmacology and Therapeutics (692) Primary Care Research (712) Psychiatry and Clinical Psychology (5448) Public and Global Health (9239) Radiology and Imaging (2202) Rehabilitation Medicine and Physical Therapy (1370) Respiratory Medicine (1196) Rheumatology (596) Sexual and Reproductive Health (714) Sports Medicine (530) Surgery (712) Toxicology (99) Transplantation (289) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a01f90652bb6fff4',t:'MTc3OTgyNzcwMg=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();