Article Text
Abstract
Background Health inequalities in cardiovascular care have been identified in the UK. The sociodemographic characteristics of patients undergoing intervention for aortic stenosis (AS) in England, and the impact of COVID-19, is unknown.
Methods National linked data sets identified all surgical aortic valve replacement (SAVR) and transcatheter aortic valve implantation (TAVI) for AS, and post-intervention cardiovascular mortality, between 2000 and 2023.
Results Of 179 645 procedures, there were 139 990 SAVR (mean age 71±10.8 years, 64% male, 96.0% white) and 39 655 TAVI (mean age 81±7.7 years, 57% male, 95.7% white). Rates of SAVR declined during COVID-19 for all groups, but TAVI rates increased steadily. Women were older; ethnic minority groups and those from most deprived areas were younger, with greater comorbidities. Women and more deprived groups had lower rates of SAVR (age-standardised rates per 100 000 in 2020–2023: 17.07 vs 6.65 for men vs women; 9.82 vs 10.10 for Index of Multiple Deprivation (IMD)-1 vs IMD-5) and TAVI (20.20 vs 9.79 for men vs women; 9.55 vs 13.36 for IMD-1 vs IMD-5). These discrepancies widened over time. Ethnic differences were observed for SAVR, with the lowest rates in black patients. Cardiovascular mortality post-intervention was lower in female patients and with decreasing deprivation, with no ethnicity-based differences.
Conclusions There are differences in intervention rates for AS in England, with lower rates in female patients and to a lesser extent, those from the most deprived areas and ethnic minority groups. These variations have widened over time. Post-intervention cardiovascular mortality is lower in women and with decreasing deprivation. Public health measures and research are needed to identify the true prevalence of AS in different populations, and the reasons for potential inequalities.
- Heart Valve Prosthesis Implantation
- Cohort Studies
- COVID-19
- Transcatheter Aortic Valve Replacement
- Aortic Valve Stenosis
Data availability statement
Data may be obtained from a third party and are not publicly available. The data used in this study are available in NHS England’s Secure Data Environment (SDE) service for England, but as restrictions apply they are not publicly available (https://digital.nhs.uk/services/secure-data-environment-service). The CVD-COVID-UK/COVID-IMPACT programme, led by the BHF Data Science Centre (https://bhfdatasciencecentre.org/), received approval to access data in NHS England’s SDE service for England from the Independent Group Advising on the Release of Data (IGARD) (https://digital.nhs.uk/about-nhs-digital/corporate-information-and-documents/independent-group-advising-on-the-release-of-data) via an application made in the Data Access Request Service (DARS) Online system (ref. DARS-NIC-381078-Y9C5K) (https://digital.nhs.uk/services/data-access-request-service-dars/dars-products-and-services). The CVD-COVID-UK/COVID-IMPACT Approvals & Oversight Board (https://bhfdatasciencecentre.org/areas/cvd-covid-uk-covid-impact/) subsequently granted approval to this project to access the data within NHS England’s SDE service for England. The de-identified data used in this study were made available to accredited researchers only. Those wishing to gain access to the data should contact bhfdsc@hdruk.ac.uk in the first instance.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- Heart Valve Prosthesis Implantation
- Cohort Studies
- COVID-19
- Transcatheter Aortic Valve Replacement
- Aortic Valve Stenosis
WHAT IS ALREADY KNOWN ON THIS TOPIC
Health inequalities in cardiovascular care have been identified in the UK, with the COVID-19 pandemic widening these gaps.
WHAT THIS STUDY ADDS
Using national linked healthcare data sets, and the national census data as the denominator data to provide age-standardised rates, we provide comprehensive assessment of sociodemographic differences and trends in rates of intervention in all patients undergoing surgical aortic valve replacement (SAVR) and transcatheter aortic valve implantation (TAVI) for aortic stenosis (AS) over the last 20 years in England.
We have identified differences in the rates of SAVR and TAVI in England, particularly for female patients and those from the most deprived areas, with these discrepancies widening over the last 20 years.
Ethnic differences were also observed for SAVR, with the lowest rates in black patients.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
There is a need to identify the true prevalence of AS in different populations, to confirm if the observed differences in rates of procedures reflect true health inequalities.
Widening health disparities call for public health measures to try and address potential inequalities, as well as research to understand the reasons for these disparities.
Introduction
Aortic stenosis (AS) is the most common valvular heart disease requiring intervention in high-income countries. Its prevalence is estimated at 12.4% in adults >75 years in Europe and North America1 and is increasing with an ageing population. A modelling study estimated severe AS in 291 448 people ≥55 years in the UK in 2019.2 The only treatment to improve prognosis is either surgical aortic valve replacement (SAVR) or transcatheter aortic valve implantation (TAVI), which is currently recommended once symptoms or left ventricular dysfunction develop.
Ethnic minority groups make up 18% of the UK population3 and the burden of cardiovascular risk factors and disease is disproportionally higher in these groups. Several studies, predominantly in the USA, have highlighted health inequalities in access to treatment for AS, with black patients being less likely to undergo SAVR and TAVI than white patients.4 Differences in pathophysiology, presentation, treatment and outcomes between sexes have been demonstrated.5 6 The association between low socioeconomic status (SES) and worse life expectancy is well established.7 However, it is unclear if there are disparities in access to interventions for AS in the UK.
The COVID-19 pandemic led to major delays in surgical procedures, and potentially widened health inequalities in the UK.8 9 National Health Service (NHS) England therefore called for urgent action to tackle inequalities and acknowledged the crucial role of high-quality data to understand the extent of the problem.
This study used national whole population linked healthcare data sets to investigate the sociodemographic characteristics and associated outcomes of patients undergoing SAVR/TAVI in England. Trends over the last 20 years and the impact of COVID-19 were examined.
Methods
Study design and participants
In this retrospective, observational population-based study, in collaboration with the British Heart Foundation (BHF) Data Science Centre and Health Data Research UK, we used the Hospital Episodes Statistics Admitted Patient Care (HES APC) database, linked to the Office of National Statistics (ONS) for death data, National Adult Cardiac Surgery Audit (NACSA) and national TAVI Audit data, and the COVID-19 General Practice Extraction Service Data for Pandemic Planning and Research (GDPPR), accessed in NHS England’s Secure Data Environment service for England via the Data Science Centre’s Cardiovascular Disease (CVD) COVID-UK/COVID-IMPACT Consortium. All patients >18 years, undergoing their first procedure for SAVR or TAVI for AS in England between 1 January 2000 and 1 December 2023, were identified using International Classification of Diseases (ICD)-10 and Office of Population Censuses and Survey’s Classification of Interventions and Procedures V.4 codes (online supplemental table 1). Data were only provisioned within the Secure Data Environment from 2013 for NACSA, 2018 for the TAVI Audit. Therefore, HES APC was used as the primary source for full cohort curation.
Supplemental material
Exposure groups
Key exposure groups were: date of birth (age), sex, ethnicity and Indices of Multiple Deprivation (IMD) as close as possible to the date of procedure. The English IMD are a composite measure of deprivation calculated for small geographical areas known as Lower-layer Super Output Areas, each representing approximately 1000–3000 people. The IMD combines scores across seven weighted domains: income, employment, health, education and skills, barriers to housing and services, crime, and living environment, using a total of 39 separate indicators. The combined scores are used to rank the areas relative to one another, identifying areas with the highest levels of deprivation. SES was based on the nearest available patient-level IMD Score, which was ranked into quintiles (most deprived (IMD-1) to least deprived (IMD-5)). Ethnicity was categorised according to the 2021 ONS Census ethnic group variables (18 categories), and then mapped to the five higher-level ethnicity categories (Asian or Asian British, black or black British, mixed or multiple ethnic groups, white and other). The primary source of ethnicity classification was HES APC, but other sources were used to supplement missing data (including GDPPR).
Covariates
Baseline cardiovascular risk factors and comorbidities were derived using HES APC ICD-10 codes, due to variable data coverage in GDPPR. These included coronary artery disease, diabetes, hypertension, asthma or chronic obstructive pulmonary disease, cerebrovascular disease, chronic kidney disease, heart failure or peripheral arterial disease.
Outcomes
The primary outcome was cardiovascular mortality post-intervention. Hospitalisation with heart failure (HHF) was considered a secondary outcome. Date and cause of death were ascertained from linked ONS mortality records. We coded deaths as ‘cardiovascular’ when an ICD-10, chapter 9 code was recorded in the primary position. Follow-up was until death or 31 December 2023.
Statistical analysis
Descriptive statistics are presented as number and proportion for dichotomous data, mean and SD for age, which was normally distributed, and median and IQR for skewed continuous data. We compared age using Student’s t-test and categorical variables using the χ2 test between different sociodemographic subgroups. Differences are considered statistically significant at p<0.05, although we acknowledge that with such a large data set, even small differences might meet this threshold, and so we have also considered the clinical relevance of any differences.
Age-standardised annual SAVR/TAVI procedure rates for different sex, ethnicity and SES groups, were estimated using the 2021 census population data as the denominator, and age-standardised to the European Standard Population data. Rates are presented per 100 000 person years.
We also calculated age-adjusted incidence rate ratios (IRRs), using a Poisson regression model, with 95% CIs, as relative risks for each of ethnicity, sex and deprivation in the different treatment periods, especially the early (2000–2004, for SAVR), prepandemic (2015–2019) and postpandemic (2020–2023) treatment periods.
For time-to-event analysis for outcomes, we used Cox proportional hazards models. We looked at the association between each of sex, ethnicity and SES with the outcome, after adjusting for age and other sociodemographic characteristics (Model-1); and adjusting for comorbidities and treatment period in addition (Model-2).
This analysis was performed according to a prespecified analysis plan published on GitHub, along with the phenotyping and analysis code (https://github.com/BHFDSC/CCU056_01).
Patient and public involvement
The project was developed in consultation with patients with cardiovascular disease and clinicians, and forms part of a larger mixed-methods project, which included a public co-applicant. The CVD-COVID-UK/COVID-IMPACT Project Proposal CCU056 was presented to their Public Contributor group and feedback taken on board prior to final approval.
Results
We identified 179 645 patients undergoing SAVR or TAVI for AS between 2000 and 2023 (table 1). This included 139 990 SAVR (mean age 70.6±10.8 years, 64% male) and 39 655 TAVI (mean age 80.8±7.7 years, 57% male). The proportions of white, Asian and black patients were 96.0%, 2.2% and 0.6% for SAVR, and 95.7%, 2.7% and 0.9% for TAVI, respectively.
Patients were more likely to be in the least deprived IMD quintiles (43.7% in IMD 4–5 compared with 32.7% in IMD 1–2 for SAVR, and 47.5% in IMD 4–5 compared with 30.1% in IMD 1–2 for TAVI). As expected, the TAVI cohort had significantly higher prevalence of most comorbidities.
The cohort by sex, ethnicity and SES
Compared with male patients (online supplemental table 2), women were significantly older (SAVR: 72.0±10.6 vs 69.8±10.9 years, p<0.001; TAVI: 81.6±7.4 vs 80.2±7.9 years, p<0.001) but had a lower prevalence of most comorbidities.
Non-white patients were significantly younger at the time of SAVR (mean age 70.8±10.7 years, 64.2±13.5 years and 65.7±13.7 years in white, Asian and black patients, respectively) (online supplemental table 3). In contrast to white patients, the majority of Asian and black patients were from more deprived quintiles (proportion in IMD 1–2 being 32.0%, 54.9% and 71.2% for white, Asian and black patients, respectively). All comorbidities apart from peripheral arterial disease were more common in Asian compared with white patients. Black patients had significantly higher prevalence of diabetes, hypertension, chronic kidney disease and heart failure, but lower prevalence of coronary artery disease, asthma/chronic obstructive pulmonary disease and peripheral arterial disease. In the TAVI cohort, Asian patients were younger (mean age 80.9±7.6 years, 77.4±9.2 years and 80.0±8.4 years in white, Asian and black patients respectively), and similar trends for SES (proportion in IMD 1–2 being 29.2%, 45.6% and 70.7% for white, Asian and black patients, respectively) and comorbidities were observed.
Those from more deprived areas were younger at the time of SAVR (age 68.1±11.7 years for IMD-1 vs 71.9±10.3 years for IMD-5) and TAVI (age 78.7±8.5 years for IMD-1 vs 81.9±7.1 years for IMD-5) (online supplemental table 4). The prevalence of almost all comorbidities was lower with decreasing deprivation.
Procedure urgency
For the subgroup of patients with the provisioned NACSA data set, we compared the urgency of SAVR (table 2). The proportion of elective procedures was similar for men and women. White patients had the highest proportion of elective procedures, followed by Asian and black patients (77%, 73% and 68%, respectively). The proportion on non-elective procedures increased with increasing deprivation (20% for least deprived, 27% for the most deprived).
Age-standardised rates of SAVR and TAVI and trends over time
Temporal trends in age-standardised procedure rates per 100 000 person years, split by sex, ethnicity and SES, are shown in figure 1 and table 3. The observed fall-off in numbers towards the end of the study period is attributed to a known lag in data capture. Specifically, although data sets are archived monthly, full coverage and record completeness typically lag by approximately 3–4 months. Therefore, the archived data do not yet reflect the full scope of records for the most recent months included in the graphs. This lag results in fewer observations towards the end of the study period, which should not be interpreted as a genuine decline in the event rates, but rather as incomplete data reporting for that time period.
The rate of SAVR was consistently higher in men compared with women, with the gap increasing over time. The overall rate declined during the pandemic, but remained higher in men (age-standardised rate 17.07 per 100 000 men compared with 6.65 per 100 000 women between 2020 and 2023). The overall rate of TAVI increased over time, with no reductions noted during the pandemic, and rates being consistently higher in men, with the gap widening over time (age-standardised rate 20.2 per 100 000 men compared with 9.8 per 1 00 000 women between 2020 and 2023).
Considering ethnicity, age-standardised rates of SAVR have been consistently highest in the white population (figure 1B). Rates increased over time in ethnic minority groups, other than black patients, in whom rates have remained consistently lowest (19.96, 15.53 and 8.99 per 100 000 white, Asian and black patients in 2015–2019, compared with 11.75, 9.45 and 4.69 per 100 000 white, Asian and black patients postpandemic). Less variations were observed by ethnicity for TAVI.
Considering SES, rates of SAVR were similar across IMD quintiles between 2000 and 2009, but differences developed from 2010 to 2019, with lowest rates in the most deprived quintile (figure 1C). For TAVI, the earlier years showed no difference in rates by IMD quintiles, although it is difficult to ascertain differences due to overall numbers being low. Variation in rates by SES was noted from 2017 onwards, with lowest rates in the most deprived group: 9.55 vs 13.36 per 100 000 population for IMD-1 vs IMD-5 in 2020–2023.
Relative risk of intervention over treatment periods
The age-adjusted IRRs were explored in the early (2000–2004), prepandemic (2015–2019) and postpandemic (2020–2023) treatment periods for SAVR and the prepandemic and postpandemic treatment periods for TAVI (table 4).
Compared with male patients, female patients had a consistently lower incidence rate of having SAVR or TAVI, across all time periods. The gap has widened over time with the IRR for SAVR being 0.57 (0.53–0.60) in 2000–2004 and 0.39 (0.37–0.42) in 2020–2023, in women compared with men. The incidence rate for TAVI has remained lower in women, but relatively stable: 0.54 (0.50–0.58) in 2015–2019 and 0.48 (0.46–0.51) in 2020–2023.
Compared with white patients, while the unadjusted incidence rates of SAVR were significantly lower in all non-white groups across all time periods, after adjusting for age, the gaps reduced over time, becoming non-significant for Asian patients in 2020–2023 (0.87 (0.72–1.04)), and mixed or multiple ethnic groups from 2015. However, the incidence rate of SAVR remained consistently lower in black compared with white patients (0.43 (0.30–0.61) in 2020–2023). For TAVI, the age-adjusted incidence rate remained higher in Asian compared with white patients in the prepandemic and postpandemic periods (1.26 (1.06–1.49) in 2020–2023), but lower in black compared with white patients postpandemic (0.73 (0.55–0.98) in 2020–2023).
There were almost no differences in the incidence rate of having an SAVR between IMD quintiles in the 2000–2004 time period. However, in the prepandemic period (2015–2019), the age-adjusted incidence rates for all IMDs were significantly higher compared with the most deprived IMD-1. These were no longer significant postpandemic as the SAVR declined across all groups. For TAVI, the incidence rate remained significantly higher across all IMD quintiles compared with IMD-1, and also demonstrated an incremental increase with decreasing deprivation, in both the prepandemic and postpandemic time periods.
Outcomes following SAVR and TAVI
Following SAVR, cardiovascular mortality was lower in women compared with men, with no difference between ethnic groups, after adjusting for age and sociodemographic factors (Model-1, figure 2). There was clear, graded association between increasing deprivation and risk of cardiovascular death. Following TAVI, cardiovascular mortality was lower for female compared with male patients, similar among ethnic groups, and significantly lower in IMD 3–5 compared with IMD 1.
Risk of HHF post-SAVR (Model-1) was similar in male and female patients, higher in Asian and black patients compared with white, and showed a graded increase with deprivation. Post-TAVI, the risk of HHF was higher in Asian compared with white patients, and with increasing deprivation, but lower in female compared with male patients (online supplemental figure 2). Results for Model-2 are shown in online supplemental figures 1,3).
Supplemental material
Supplemental material
Supplemental material
Discussion
Using large, linked national data sets, we provide a comprehensive overview of SAVR and TAVI procedures for AS over the last 20 years for the whole population of England. We describe the sociodemographic characteristics, trends over 20 years, impact of COVID-19 and post-intervention outcomes. The most prominent differences in procedure rates were for female patients and those from the most deprived areas, with these discrepancies widening over time. Ethnic differences were observed for SAVR, with the lowest rates in black patients. However, cardiovascular mortality following SAVR/TAVI was lower in female patients, higher with increasing deprivation and not different among ethnic groups.
Patient characteristics
Female patients were significantly older with fewer comorbidities. This may reflect later presentation, or delays in diagnosis and/or referral in women. A higher prevalence of discordant AS in women may contribute to this, though the possibility of AS developing at a later age cannot be excluded. Sex differences in anatomical features (less calcification), remodelling response to AS and less referral for SAVR have previously been reported.5 10
Ethnic minority patients were younger at the time of intervention, with more comorbidities. A greater proportion were from the most deprived quintiles, which is reflective of the general population in England, highlighting the importance of ‘intersectionality’ in healthcare access, and the difficulty in looking at a specific sociodemographic characteristic in isolation. Studies from the USA have suggested a lower prevalence of AS in black, Hispanic and Asian patients,11 but these observed differences likely reflect lower access to echocardiography and healthcare, rather than biological differences in disease occurrence. This is supported by the younger age of black and Asian patients at the time of SAVR in our study. There is some conflicting data on the prevalence of bicuspid aortic valve being higher in Asian patients, which may result in earlier intervention. While the true prevalence of AS in the UK is not known, in the OxValve Study,12 undiagnosed AS was twice as prevalent in the two most deprived SES quintiles. OxValve was conducted in a relatively affluent and predominantly white population (99%), and did not look at ethnic differences.
Contributing factors may include lower rates of presentation to primary care and/or referral for imaging and treatment, which is also suggested by the greater incidence of non-elective SAVR in the black cohort, hinting at possibly later presentation/diagnosis. Urgent procedures are associated with worse outcomes, with the latest national TAVI audit report showing a higher inpatient (1.5% vs 0.6%) and 30-day (3.1% vs 1.3%) mortality in those undergoing an urgent versus elective TAVI.13 However, the overall long-term cardiovascular mortality was not significantly different between ethnic groups in our study, although HHF was more common.
Rates of SAVR
While sociodemographic disparities in treatment of AS have been described in the USA, differences in population demographics and healthcare service preclude direct comparison with our data. A publication in 202314 summarised the current state of knowledge on cardiovascular disease inequalities in care delivery and outcomes in the UK, including TAVI, and concluded that there are ‘limited publications about ethnicity in the UK’ but inequitable cardiovascular care within the NHS based on age, sex and geography.
Two published studies15 16 specifically looked at sociodemographic differences in patients undergoing SAVR and TAVI in England. In a single tertiary centre over 5 years (2017–2022),15 while the crude rates of interventions were lower in South Asian patients, the population age-standardised rates were not statistically different. In a summary of HES data over 3 years,16 female patients and those from an ethnic minority and lower SES had a lower ‘odds’ of having an aortic valve intervention. However, the ‘denominator’ in the aforementioned study consisted of only those with an inpatient coded diagnosis of AS within the HES database, and not the general population, which does not consider the full population ‘at risk’. This is relevant as most people with AS are diagnosed and monitored as an outpatient. Crucially, they also did not provide age-standardised results, which is a limitation, as there are known differences in the age distribution of different ethnic groups within the UK. Additionally, the results were not split by SAVR and TAVI, which have different demographics and outcomes.
Post-intervention outcomes
Despite differences in rates of SAVR, post-intervention cardiovascular mortality rates were similar among ethnic groups. Studies in the USA have confirmed similar post-SAVR survival between black and white patients, despite lower rates of SAVR in black patients,17 and no differences in outcomes between ethnic groups was confirmed in a recent systematic review.4 Female patients had significantly lower cardiovascular mortality despite having lower rates of intervention and being older at the time of intervention. There are conflicting data on sex differences in post-SAVR outcomes. While studies have shown higher post-SAVR mortality in women, these differences were not significant once confounding factors were corrected for. Registry data from the USA and Germany confirm lower mortality in women post-TAVI. Lower SES class was associated with worse outcomes. Although no contemporary UK national data are available, a previous UK registry database of primary aortic and mitral valve replacement between 1986 and 2001 also showed the association of deprivation with lower survival,18 and an analysis of the national cardiac surgery database between 2003 and 2013 confirmed higher in-hospital and mid-term mortality in those from the most deprived SES quintile.19 SES is known to be a predictor of postoperative mortality following elective surgery20 and our data confirm this for SAVR and TAVI in England.
Strengths and limitations
This comprehensive study of the sociodemographic characteristics of those undergoing intervention for AS for the whole population in England, using real-world linked data sets, provides an accurate assessment of those receiving treatment for AS over 20 years. Using real-world data also has its limitations, as the quality of the data is dependent on the quality of coding, which can result in missing data. However, we have no reason to suspect systematic bias in this regard between SAVR and TAVI. This was mitigated to some extent by using multiple linked data sets, allowing more complete data. An important limitation is lack of echocardiographic data in the treatment cohort and lack of true AS prevalence data in the UK population, making it impossible to know if there are differences in AS prevalence or AS severity between subgroups.
Conclusions
There are differences in rates of SAVR and TAVI in England, with lower rates in female patients and to a lesser extent, those from the most deprived areas, and these discrepancies have widened over time. Ethnic differences were also observed for SAVR, with the lowest rates in black patients. Public health measures and research are needed to identify the true prevalence of AS in different populations, and the reasons for potential inequalities.
Data availability statement
Data may be obtained from a third party and are not publicly available. The data used in this study are available in NHS England’s Secure Data Environment (SDE) service for England, but as restrictions apply they are not publicly available (https://digital.nhs.uk/services/secure-data-environment-service). The CVD-COVID-UK/COVID-IMPACT programme, led by the BHF Data Science Centre (https://bhfdatasciencecentre.org/), received approval to access data in NHS England’s SDE service for England from the Independent Group Advising on the Release of Data (IGARD) (https://digital.nhs.uk/about-nhs-digital/corporate-information-and-documents/independent-group-advising-on-the-release-of-data) via an application made in the Data Access Request Service (DARS) Online system (ref. DARS-NIC-381078-Y9C5K) (https://digital.nhs.uk/services/data-access-request-service-dars/dars-products-and-services). The CVD-COVID-UK/COVID-IMPACT Approvals & Oversight Board (https://bhfdatasciencecentre.org/areas/cvd-covid-uk-covid-impact/) subsequently granted approval to this project to access the data within NHS England’s SDE service for England. The de-identified data used in this study were made available to accredited researchers only. Those wishing to gain access to the data should contact bhfdsc@hdruk.ac.uk in the first instance.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. The North East - Newcastle and North Tyneside 2 research ethics committee provided ethical approval for the CVD-COVID-UK/COVID-IMPACT research programme (REC No 20/NE/0161) to access, within secure trusted research environments, unconsented, whole-population, de-identified data from electronic health records collected as part of patients’ routine healthcare. The study protocol was approved by the Approvals & Oversight Board of the BHF Data Science Centre’s CVD-COVID-UK/COVID-IMPACT consortium (project CCU056).
Acknowledgments
The authors thank all data providers who make health relevant data available for research.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @DrAnvesha_Singh
Contributors AS, SA, IS, KK, GPM and CL contributed to the conception and design of the work. FC, TB and ASt contributed to the acquisition and analysis of the data, and along with AS and CL to the interpretation of the results. AS drafted the initial manuscript and all co-authors revised it critically for important intellectual content. All authors gave final approval of the version to be published and agree to be accountable for its content. AS is responsible for the overall content of the manuscript as the guarantor.
Funding The British Heart Foundation Data Science Centre (grant No SP/19/3/34678, awarded to Health Data Research (HDR) UK) funded co-development (with NHS England) of the Secure Data Environment service for England, provision of linked data sets, data access, user software licences, computational usage, and data management and wrangling support, with additional contributions from the HDR UK Data and Connectivity component of the UK Government Chief Scientific Adviser’s National Core Studies programme to coordinate national COVID-19 priority research. Consortium partner organisations funded the time of contributing data analysts, biostatisticians, epidemiologists and clinicians. In addition, a British Heart Foundation Data Science Centre and Health Data Research UK rapid funding call (HDRUK2023.0267) was awarded to A Singh, for end-to-end service, including data curation and analysis, to complete this work.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.