Barts structural Interventional Registry (BSIR)

The aims of this research are:

- To understand the impact of disease and treatments on the mortality of patients with valvular heart disease.

- To compare demographic and clinical characteristics, biomarkers, therapies, their complications on quality of life, hospitalisation and mortality among patients with valvular heart disease.

- To validate simulations of procedures which have been undertaken in real patients and to correlate numerical results with retrospective clinical measurements.

The data obtained from NHS digital will directly contribute towards achieving some of these aims.

Bart's Health NHS Trust are the Sole data Controller. They are the sponsor for the study. Bart's Health NHS Trust encompasses several hospitals including Royal London Hospital and St Bartholomew's Hospital. Bart's Health Centre (dept. of Cardiology) where the cohort data originates from, is part of St Bartholomew's Hospital.

Queen Mary University is affiliated to Bart's Health NHS Trust. Queen Mary University run the Joint Research Management Office for Bart's Health NHS Trust and provide a governance function only to make sure the study protocol is adhered to. Queen Mary University will have no involvement in the decision making or processing of NHS Digital data.

Rationale for this research:

Valvular heart disease (VHD) affects a significant proportion of the global population. With advances in technology, treatments and Bart's understanding of cardiac disease, Barts Heath NHS Trust can now treat VHD using new therapies. However, the trials that have introduced these therapies have strict inclusion and exclusion criteria, resulting in many patients being excluded. Consequently, there is a great deal of uncertainty about the safety and efficacy of treatments for many patient populations. Real-world evidence of treatments among patients with VHD is increasingly recognised as an important part of providing safety and efficacy data and improving the care that Barts Heath NHS Trust provide. Additionally, most trials include hard end-points such as mortality, but few take into consideration symptomatic relief and quality of life- factors that are just as important for many patients. This study will fill the gap in knowledge by providing answers to questions patients want to know, such as ‘will this treatment let me walk again without breathlessness?’

Computer models have been shown to replicate accurately anatomical and physio-pathological conditions and to predict acute outcomes of treatments in specific cases of patients with VHD. In-silico methods can potentially be used to assess the results of procedures on entire populations. Hence, so-called in-silico trials can contribute to reduce the uncertainty and explore, with synthetic data, benefit or risks of specific procedures with no harm to the patients. A large scale validation of such methods on VHD data is however still missing. This study will enable validation of methods and identification of crucial correlations between simulations, imaging and clinical findings and real world outcomes.

VHD is seldom caused by an individual disease process or risk factor. More commonly, it is a consequence of several pathological variables. Therefore in order to understand VHD, it is important to study all the relevant risk factors and disease processes involved. For example, aortic stenosis- narrowing of a heart valve coexists with coronary artery disease and both share the same risk factors and pathophysiological mechanisms. This study will allow the study of both these conditions and several other combinations by its encompassing nature.

Overall study population is approximately 10,000 people. These patients are largely elderly with a mean age of 75 years. Bart's Health NHS Trust serves a large ethnic minority population- Asian and afro-Caribbean but also Caucasian. Patients are roughly split 1:1 male to female. The population will be studied in its entirety.

This study includes patients with all strata of VHD severity as Barts Heath NHS Trust now know that even less than severe VHD has clinical implications. Understanding the biology, natural history, impact of medical and interventional treatments in these patients is important. This will potentially help understand which patients need earlier treatment, which require closer follow up and how to risk stratify patients according to clinical characteristics.

COVID-19 has changed the way healthcare systems operate, with most patients now being cared for locally rather than coming to the Barts Heath NHS Trust centre. As a result Barts Heath NHS Trust healthcare staff know less about treatment success and failure, complications, hospitalisations and mortality than they used to. This has hampered important progress in the care provided to patients. By obtaining outcome data with this study, Bart's Health NHS Trust will be able to understand what works and what doesn’t work for particular types of patients.

This study will add to that void in literature by assessing the characteristics and outcomes of patients with VHD. It will also identify predictors of outcomes, improve risk stratification and diagnostic evaluation. Essentially the clinical trials that have led the field have highly selected patient populations such as those with a single valvular disorder and the absence of complex coronary artery disease. We essentially want to answer questions such as: how do we manage patients with complex coronary artery disease and coexist any valvular heart disease. Using the data from NHS digital we can answer this by understanding if we don’t treat the coronary artery disease the proportion of patients that have a heart attack or heart failure or die are xxx whereas if we do treat the coronary artery disease the same outcomes occur in yyy patients. Similarly with regards to multiple valvular heart disease we want to understand if we fix all the valve problems the rate of heart failure and death is xxx whereas if we fix only the main one then the same rate is yyy. We hope to answer these questions and publish our findings in peer-review journals and present our findings at international conferences.

Risks to patients

There is no expected medical risk for any patient related to the current study. This study is an observational, retrospective study and does not require any specific clinical intervention or procedures (e.g. physical or psychological examinations or tests); no drug is given to the patient. Patient identifiable data will only be handled by the clinical team and pseudonymised before the research team uses the data. Only the minimum dataset required will be used at any point to answer a specific research question. Data transferred to and from NHS digital will be a minimum dataset of identifiable patient data, which will be encrypted and sent through secure, well-established channels following strict protocols. Permission to use identifiable data without patient consent has been approved by the Confidentiality Advisory Group (CAG). These protocols will prevent the risk of confidentiality breaching and unintended disclosures of patient data.

Bart's Health NHS Trust are requesting HES APC data to study how certain clinical, demographic and imaging features affect rehospitalisation, further procedures and complications. Civil registration data is requested to evaluate what patient characteristics are associated with mortality and to establish the cause of death. Only data that is essential to answer the study questions is being requested from NHS Digital. A specific cohort from a specific geographical location during a specific time period will be submitted.

The legal basis for processing data under article 6 (1) (e) and article 9 (2) (j) of GDPR. The reason this study is of public interest is because valvular heart disease is common affecting 1 in 2 adults over the age of 65. Barts Heath NHS Trust need to improve the care that Bart's provide to these patients by optimising the timing of treatments, identifying factors associated with poor and good outcomes and refining clinical pathways. The findings from this study could potentially provide the basis for such improvements. Bart's Health NHS Trust serves a large ethnic minority population- Asian and afro-Caribbean but also Caucasian. Ethnic minorities are largely excluded from the current literature making this study very important for the public by filling an important gap in the literature.

Interest to public and patient/public involvement (PPI)

This research is anticipated to be of great interest to the public as it will provide answers to some of the important but little studied areas of valvular heart disease. The large number of patients in this study will mean that Barts Heath NHS Trust will be able to generate significant real-world data. Additionally Bart's Health provides care to a large population of ethnic minorities (of African and Asian heritage). These ethnicities are often less presented in clinical trials, making this study even more important. The trials included a predominantly Caucasian population with little representation from Asian or afro-Caribbean patients due to geographical reasons- most trials took place in Europe and USA.

PATIENT INFORMATION POSTER

A PPI study conducted by Bart's also identified the need to conduct research such as this study as it provides real world data and includes many patients that would otherwise be excluded from clinical trials.

The PPI study also gave insight into what is important for patients and often not assessed in clinical trials, such as quality of life and symptoms.

These factors have played a fundamental role in the design of this study

In order to ensure that this research activity doesn’t cause harm to patients and complies with data protection laws, obtained approval for this study has been obtained from the Bart's Data Protection Service, Research Ethics Committee, the Joint Research Management Office at Bart's/Queen Mary University and the Confidentiality Advisory Group.

Bart's Heart Centre have been conducting clinical research for several years and have accumulated a large database of patients with heart disease. This study has been derived from previous work that started in 2018 and included both retrospective and prospective observational studies. These studies have highlighted significant areas of valvular heart disease that were not previously known and therefore made a significant impact on patient care.

Patients with valvular heart disease are divided according to type of valve disease which include:

- aortic stenosis

- mitral stenosis

- aortic regurgitation

- mitral regurgitation

- tricuspid regurgitation

and by severity:

- mild

- moderate

- severe

The purpose of this project is:

- To understand the impact of disease and treatments on the mortality of patients with valvular heart disease.

- To compare demographic and clinical characteristics, biomarkers, therapies, their complications on quality of life, hospitalisation and mortality among patients with valvular heart disease.

- To validate simulations of procedures which have been undertaken in real patients and to correlate numerical results with retrospective clinical measurements.

Outcome data for Bart's patients including:

- admission to hospital for: heart failure, myocardial infarction, coronary revascularisation, stroke, haemorrhage, arrhythmia

- date and duration of admission to hospital for: heart failure, myocardial infarction, coronary revascularisation, stroke, haemorrhage, arrhythmia

- details regarding any treatment delivered during these admissions: e.g. revascularisation performed using percutaneous coronary intervention or coronary bypass grafting, 2 stents or 3 bypass grafts, etc

- date of death

- cause of death

The data is required from 2015 to the present because Bart's started collecting data for patients from 2015. Only including a certain proportion of patients will lead to recruitment bias and confound Bart's analysis and outcomes. There is no other alternative or less intrusive method of obtaining this data. This data will be handled by substantive employees of Bart's Health NHS Trust that have training and access to confidential patient information. Any data that will be used for research purposes will be pseudonymised before doing so. The data will be pseudonymised before it is analysed. Bart's cannot narrow the data by geography or demographics as this will lead to selection bias and skew the results. Bart's have limited data to patients with valvular heart disease. Not all patient episodes are required for this research, hence why data from NHS digital regarding the most important and the bare minimum that is required to conduct this research, has been requested. Specific dates are required for this research as this will have significant impact on the results of this research. Bart's need to accurately identify if an outcome occurred within given time frames of a certain treatment.

Bart's Heath NHS Trust have taken steps to reduce the amount of data that Bart's send to NHS digital and that Bart's request from NHS digital. These datasets are the minimum necessary to fulfil the research aims, primarily to determine the safety and efficacy of treatments offered.

There are no external organisations involved in this research. Bart's Health NHS Trust is the Data Controller who processes the data. Data from NHS digital will only be used by substantive employees of Bart's Health NHS Trust.

Barts Health NHS Trust will submit to NHS digital the minimum identifiable data that NHS digital require in order to identify patients and provide their relevant outcome data. This includes name, date of birth and NHS number.

The Study will receive from NHS digital the date and duration of hospitalisation for certain conditions and mortality and also obtain the date and cause of death.

In order to reduce the number of people who have access to identifiable data, availability will be restricted to clinical staff who already have access to identifiable patient data.

The clinical team will comprise of 4 individuals who are substantive employees of Barts Health NHS Trust. The research team will comprise of these 4 clinical staff and an additional 3 substantive employees of Barts health NHS Trust.

Any data that is used for research will be pseudonymised and analysed. The data (both actual and pseudonymised will remain at Barts Health NHS Trust.

The data obtained from NHS digital will be linked using a study ID to the local database at Barts Health NHS Trust that contains details on the patient demographics, comorbidities, imaging and procedural details. The data from NHS digital will not be linked to any other data. This data linkage has been approved by the Research Ethics Committee, Confidentiality Advisory Group, Joint Research Management Office and Barts data protection service.

In order to prevent reidentification of the patient data it will be pseudonymised before being used for analysis. When data is required for research, a query will be raised, and a search performed by the clinical team. This data will be pseudonymised before being used by the research team. Pseudonymisation will be performed by assigning each patient with a study number. A separate linkage file will be held by the CI within a restricted drive on the NHS server at Bart's Health providing information on the study number and its corresponding identifiable patient information. Data that is used for research will only include the study number, without patient names, date of birth or hospital ID. The non-identifiable data that NHS digital send back (Study ID, date of death and diagnosis information) will be matched with Bart's local database that contains clinical, demographic and procedural data and specific study IDs. The Study ID provided by NHS Digital to Bart's should match that held in Bart's database and will allow for linkage. Bart's will not re-access the original dataset that is held within the NHS servers to re-identify the data.

All data including the pseudonymised research data will be stored on the NHS clinical server ensuring security of the data. There will not be any attempt to re-identify patients.

The CI for this study will ensure and take responsibility for all persons that have access to this data will be appropriately and regularly trained in data protection and confidentiality, and no access to the data will be granted until this training has been completed. Bart's Health NHS Trust will ensure that only persons to have completed the Research ethics and UK General Data Protection Regulation (UK GDPR) training (https://www.ed.ac.uk/literatures-languages-cultures/research/ethics/research-ethics-general-data-protection-training) will have access to this data and data access will be restricted to a few named individuals. All people with access will be substantive employees of Bart's Health NHS Trust.

All data will be stored in secure servers at Bart's Health NHS Trust. The data will be password protected and access restricted to named individuals. The data will not leave Bart's Health NHS Trust.

The cost of acquiring this data from NHS digital will be paid by an unrestricted research grant that is held by the Medical Doctor and Consultant at Bart's Health NHS Trust. This grant was provided by Edwards Lifesciences, who will not have anything to do with how the data from NHS digital is used/processed. They are not funding this research study, neither will they have any involvement in the research that Bart's Heath NHS Trust conduct. They have simply provided Bart's Heath NHS Trust with a grant and Bart's Heath NHS Trust are using some of that money to make this application to NHS digital.

This study will want to fill the void in literature by assessing the characteristics and outcomes of patients with VHD. It will also identify predictors of outcomes, improve risk stratification and diagnostic evaluation.

The output of this research will be presented at local, national and international meetings and conferences, these will include but not be limited to the European Society of Cardiology Congress, PCR (percutaneous coronary interventions) London valves and TCT (transcatheter cardiovascular therapeutics). It will also be published in peer-reviewed journals, including but not limited to European heart journal, JACC and Circulation. Dissemination of the output will be facilitated by social media, news, teaching courses and webinars. Barts Heart Centre has a strong affiliation with media channels and will communicate the findings widely.

The output from the use of NHS digital data will include identifying clinical and imaging predictors of mortality, cardiovascular mortality, stroke, heart failure hospitalisation and coronary revascularisation.

The Study aims to produce outputs from its work at the end of the first year of obtaining data from NHS digital in Q3 2023.

Study findings will be shared with the patients and the public via the study website. All data shared will be aggregated with small numbers supressed in line with the HES Analysis Guide.

The dissemination of the output from this research is hoped it will fill the void in literature by assessing the characteristics and outcomes of patients with VHD. It will also aim to identify predictors of outcomes, improve risk stratification and diagnostic evaluation. There is a paucity of data relating to specific subgroups of patients with valvular heart disease that have not been included in clinical trials. This includes research involving ethnic minorities. Bart's Health serves a large population of Ethnic minorities (of African and Asian heritage). This research aims to fill this void by providing real World evidence.

This research will involve a large cohort of patients from one of Europe’s busiest cardiac centres. The output has the capability of transforming Barts Heath NHS Trust clinical practice by improving diagnostic and treatment pathways, providing better evidence for risk stratification such that care is delivered to those who need it the most in a timely and effective method.

The findings of this study is hoped to change how Barts Heath NHS Trust treat patients with valvular heart disease, when Barts Heath NHS Trust intervene, and what the best management strategy is. These changes are hoped to lead to improvements in the quality of life, symptoms, mortality and heart failure hospitalisation. If the findings of this study identify areas of improvement in care, risk stratification, timing of intervention or type of treatment, Bart's will be better able to inform patients, instigate changes in management and optimise treatments. These changes could reduce hospital admission and length of stay and mortality, improve quality of life and symptoms. Overall these changes could lead to reductions in healthcare related costs and provide efficiency gains. These benefits will be achieved by healthcare systems not only in the UK but also internationally. These improvements can be measured using hard end-points such as mortality, hospitalisation, length of hospital stay and are likely to be done so over the next 5-10 years.

The output from this research will potentially change the management of valvular heart disease and thereby affect patients both in the UK and internationally. All patients with valvular heart disease stand a chance to benefit from these potential changes.

Yielded Benefits is not a requirement for new applications.