NHS Digital Data Release Register - reformatted

Sandwell and West Birmingham Hospitals NHS Trust

Project 1 — DARS-NIC-331733-T2K1Z

Opt outs honoured: No - data flow is not identifiable (Consent (Reasonable Expectation))

Sensitive: Sensitive

When: 2020/10 — 2020/10.

Repeats: One-Off

Legal basis: Health and Social Care Act 2012 – s261(2)(c), Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii)

Categories: Anonymised - ICO code compliant

Datasets:

  • Civil Registration - Deaths
  • Demographics

Objectives:

The Birmingham and Black Country Atrial Fibrillation study is a prospective longitudinal observation study run by the University of Birmingham and which started recruitment in September 2014 and finished recruitment in February 2018. This study aims to improve understanding of Atrial Fibrillation (AF). AF is a condition affecting 3% of the population in which the heart rate becomes irregular and in some cases is very fast. A very important complication of AF is heart failure. Heart failure is a condition where the pump function of the heart is reduced. As the heart is not pumping effectively in this condition, it can result in a build-up of fluid in parts of the body where fluid does not normally collect, such as the lungs and under the skin; a condition called "oedema". Heart failure can also be associated with a significantly reduced life expectancy. Predicting which patients with AF will develop heart failure and its complications is very difficult. This study, therefore, aims to develop a risk score that doctors can use to predict the risk of heart failure using a range of blood tests called "biomarkers". Biomarkers are already used routinely in other areas of cardiology. A good example of this is a biomarker called troponin. This biomarker is used by hospital doctors as a diagnostic test for a heart attack. This study, therefore, aims to determine if factors such as biomarkers collected when patients were initially recruited to this study can be used to predict outcomes of AF including cardiovascular death. This is that it would enable doctors to change patient's condition management by using medications known to prevent the progression of heart failure and its complications. The data requested is for patients in the cohort who have died and the patient's certified cause of death. By linking baseline information based on biomarkers to outcomes such as death secondary to heart failure, it will be possible to determine if biomarkers can be used to predict this outcome. This gives information on the prognostic benefit of biomarkers. By using statistical analysis, it will be possible to generate a risk prediction model that can be used by doctors in the future to predict the risk of patients with AF developing heart failure and thereby enabling doctors to modify treatment accordingly. A total of 1,630 patients referred to the Sandwell and West Birmingham Hospitals NHS Trust (SWBH), for inpatient or outpatient evaluation of acute illnesses were recruited to the study between September 2014 and February 2018 for the Birmingham and Black Country Atrial Fibrillation Registry (BBC-AF). 759 of these patients had diagnosed AF confirmed by ECG while the control group had at least two CHA2DS2-VASc risk factors (the most commonly utilized method to predict thromboembolic risk in atrial fibrillation stroke). Patients who did not have diagnosed AF underwent 7-day ambulatory an electrocardiogram (ECG) monitoring to detect silent AF. Clinical information was collected from a detailed interview, review of electronic patient records, and chart review. ECG and transthoracic echocardiography were performed in all patients. Also, blood samples were taken on all patients measuring 14 different biomarkers. This study complied with the Declaration of Helsinki, was approved by the National Research Ethics Service Committee (BBC-AF Registry, West Midlands, UK, IRAS ID 97753), and sponsored by the University of Birmingham. All patients provided written informed consent. For the purpose of this application, only the 1,487 cohort members recruited between June 2015 to February 2018 using the version 7 consent materials will be used. EU FUNDING AND THE CATCH ME PROJECT The baseline data from the BBC AF study including the plasma biomarkers already feed into the CATCH ME project (Characterising AF by Translating its Causes into Health Modifiers in the Elderly, funded by EU horizon 2020) which aims to improve the prevention and treatment of atrial fibrillation and its complications. The CATCH ME consortium is made of the following institutions; the University Of Birmingham, Consorci Institut D'Investigacions Biomediques August Pi I Sunyer, Universiteit Maastricht, Ludwig-Maximilians-Universitaet Muenchen, The Chancellor, Masters and Scholars of the University Of Oxford, Universite Pierre Et Marie Curie, Kompetenznetz Vorhofflimmern E.V and the European Society of Cardiology. Baseline data from BBC AF has been used to contribute to ongoing research into the use of biomarkers in the diagnosis of AF. However, there are no proposed plans for wider collaboration or associated work based on the outcome data from this cohort. This project will be solely managed and delivered by the University of Birmingham. Therefore, the CATCH ME project has no direct control over the BBC AF project and outputs derived from NHS Digital will not be shared with the CATCH ME consortium. The European Network for Translational Research in Atrial Fibrillation (EUTRAF) is the full title of the EU funded project. Network funding was launched in November 2010 for multiple AF related projects across Europe with active funding ending in October 2015. The grant awarded to the University of Birmingham has been used to study genetic predisposition to AF. This work has been performed by an experimental scientific group at the University of Birmingham to study molecular mechanisms of myocardial disease in transgenic animals with a focus on ventricular and atrial arrhythmia. Basic science research using mice with heterozygous deletion of one PITX2 gene (PITX2+/- mice) has been performed to assess the development of AF and shortening of the atrial action potential without obvious other structural atrial disease. The outcome data from this trial and mortality data requested from NHS Digital does not fall within the remit of European Network for Translational Research in Atrial Fibrillation grant and is as such outside the scope of this funding. The outcome data from this trial and mortality data requested from NHS Digital does not fall within the remit of European Network for Translational Research in Atrial Fibrillation grant and is as such outside the scope of this funding. Two year-follow-up appointments have been arranged for all 1,487 cohort members recruited to this study and this will finish in February 2020. However, the final analysis of this study will require linking the baseline findings of this study to outcomes in order to determine the predictive power and potential utility of biomarkers in clinical practice. It is therefore very important to obtain information on mortality and cause of death for all patients included in this study. This agreement therefore only requests data to identify which patients recruited to this study have died (Demographics extract) and for these particular patients, their date of death and cause of death from the Civil Registration (Deaths) extract. To match each outcome to the appropriate study participant, SWBH NHS Trust will submit a list of identifying details for the 1,487 cohort members (NHS Number, Date of Birth and Post Code) along with the unique Study ID assigned to each individual. The Study ID will be sufficient to link the data received back from NHS Digital at patient level to the data already held by SWBH NHS Trust. As such, it will be possible to just receive data from NHS Digital containing only the study ID and not other personal identifying information, thereby helping to minimise data dissemination. DATA MINIMISATION For data minimisation, data is restricted to only those participants in this cohort who are deceased as well as patients who are lost to follow up to rule out immortals/study bias. In order to accurately define the right cohort individuals, SWBH NHS Trust will provide the minimum information required to achieve this. Data provided will include the patient’s Study ID, NHS Number, Date of Birth and Post Code. This will facilitate cohort identification but at the same time utilise the least intrusive means to achieve this outcome. Data will be kept for the minimum time frame required to perform analysis before destroying this data in line with NHS Digital guidelines. Following consultation with NHS Digital, a period of 2 years was recommended to ensure that all analysis can be performed but also enable any queries to be addressed following peer review and publication. All participants included in this study were recruited from SWBH NHS Trust. Also, the two-year research clinic follow-up for each of the participants recruited to this study has taken place at SWBH NHS Trust. Moreover, the chief investigator for this study also works as a consultant cardiologist at SWBH NHS Trust. Therefore, to ensure the ongoing maintenance of patient confidentiality, all data obtained from NHS Digital will only be stored at this NHS facility. Obtaining this data from NHS Digital is very important as many of the patients recruited to this study attend a variety of NHS facilities and use a variety of different bereavement services. Therefore, accessing mortality data from a centralised service such as NHS Digital provides the most robust mechanism of obtaining a full data set with respect to the certified cause of death. NHS Digital data received under this data sharing agreement, including data from which identifying details have been removed, may not be shared with any third parties nor used for any additional purposes other than those specified in this Purpose section. DATA CONTROLLERS The joint Data Controllers are the University of Birmingham and SWBH NHS Trust. All patients included in this study were recruited from SWBH NHS Trust meaning that this hospital trust already manages the hospital records for each of these patients. In addition, as this is an NHS facility, all sensitive patient data is managed in line with national NHS guidelines concerning confidentiality. The Chief Investigator for this study is also Director of the Institute of Cardiovascular Sciences at the University of Birmingham, Chair in Cardiovascular Medicine and a Consultant Cardiologist at SWBH NHS Trust. The BBC AF study is co-ordinated and sponsored by the University of Birmingham. However, data is recorded, organised, structured and stored by researchers who are substantive employees by SWBH NHS Trust. Therefore, SWBH NHS Trust will be ultimately responsible for data processing for this NHS Digital application. LEGAL BASIS The General Data Protection Regulation Article 6 (1) (e) and Article 9 (2) (j) are the legal basis for the processing of the data. Article 6 (1) (e) states that processing is necessary for the public interest. The purpose of this application is to determine which patients included in this cohort study have died and also their date of death and certified cause of death. Using this data, it will be possible to determine if factors such as biomarkers can predict death from cardiovascular causes such as heart failure. This data will not be used for commercial purposes, will not be provided in record level form to any third party and will not be used for direct marketing. Only patients residing in England and Wales have been included in this study. The public interest in carrying out this study is that the results will help improve patient care by enabling doctors to predict what patients with AF will develop heart failure. This will thereby enable doctors to change condition management by allowing the early introduction of medications known to prevent the progression of heart failure and its complications.

Yielded Benefits:

Baseline data on biomarker levels in patients with and without AF who were recruited to the BBC AF study has already been combining several data sets of large clinical trials conducted within the consortium to fulfil the objective of defining mechanisms of AF aiming specifically to deliver new biomarkers and new therapeutic targets for AF.

Expected Benefits:

This research project will identify and validate combinations of clinical features, biomarkers, and potentially genetic markers that characterise patients with AF at risk of heart failure and related complications. The results will help guide the condition management of AF in patients at risk of significant cardiovascular complications. In the same way that the CHA2DS2VASC score is used to determine patients' risk of developing a stroke and to guide treatment decisions regarding anticoagulation, the identification of patients at a high risk of heart failure and its related morbidity and mortality would facilitate a more personalised approach to the medical condition management of these patients. A good example of this would be the early introduction of evidence-based treatments for heart failure, such as angiotensin-converting enzyme (ACE) inhibitors in patients identified as being at high risk of heart failure-related hospitalisation and/or mortality. This study has the potential to improve the condition management of patients with AF by enhancing the ability of doctors to identify patients with AF at risk of developing heart failure and it's complications. Using this information, this will enable doctors to initiate preventative heart failure therapies early in order to prevent the progression of this condition. To achieve the stated purposes, study findings will be submitted to peer review journals for publication. Changes in the condition management of AF are largely driven by national and international guidelines such as those published by the European Society of Cardiology. These guidelines are in turn largely driven by evidence from observational studies and randomised control trials. Therefore, the publication of this study in high impact journals and presentation at national and international conferences will be the first step in this process. Key international meetings such as the European Society of Cardiology’s annual conference is attended by key clinical policymakers, clinicians and academics. Presentation of results at this conference will target key members of the scientific community involved in developing international guidelines for the condition management of AF thereby maximising the impact of these results in improving patient care. There are many potential benefits to performing this study and disseminating the results of this study. Predicting risk of heart failure in patients with AF has the potential to improve patient care by enabling doctors to start heart failure therapies before this disease progresses further. A very important objective of all researchers involved in this study has been to communicate with the patient group after recruitment to this study. As such, patients have been sent written communication at two years following recruitment to this study to update and express gratitude to all patients for their involvement in this study. In addition, patients received a two-year follow-up appointment inviting them to attend an open nurse-led follow-up appointment which encompassed a cardiovascular health check-up. Patient feedback regarding this follow-up was very positive as all patients were offered a health screen in terms of blood pressure check, ECG to check for AF and medication review. Any health concerns raised during these appointments triggered a local review by a cardiology specialist doctor and when appropriate, contemporaneous communication and correspondence to the patient and the patient’s General Practitioner to update and advice on any medical or health promotion advice that resulted from this consultation. The research lead will be essential to the successful delivery of the project. This research fellow has already been awarded a one year Birmingham Health Partners Fellowship and has been using this time to complete the follow up of patients recruited to the BBC AF study. During this time, he has gained a thorough understanding of the parameters and variables contained within this database. The research lead has completed an MSc with Imperial College London, and during this time developed competency in medical statistics. He is therefore in a unique position to use his skill to lead and coordinate the complex analysis by serving as a bridge between cardiovascular and data science departments within the University Of Birmingham and SWBH NHS Trust for which he is also a substantive employee. On completion of this project, the research fellow's career goals are to use the knowledge and skill obtained by working on this project to apply for a PHD fellowship in 2022.

Outputs:

The BBC AF study has already been used to develop a biomarker model that can be used to predict patients with atrial fibrillation based on the baseline data obtained in this study. This study based on baseline data has already been published in the European Journal of Cardiology and presented as a poster at the European Society of Cardiology Conference. The mortality data requested from NHS Digital will be used for a follow-up study which looks at predictors of heart failure in patients with AF. The research team also plans to submit the results of this follow-up study to peer-reviewed journals and presentation at national and international conferences. All output data generated from this study will be anonymised and will be presented using standard research software. Statistical packages such as SPSS will be used to analyse data and output will be presented as graphs using this software. For example, a Kaplan Myer curve (estimator is used to estimate the survival function) will be used to demonstrate mortality rates for all patients included in this study. All outputs will be aggregated data with small numbers suppressed. In order to disseminate important findings from this study, results will be submitted to international scientific journals for peer review and publication. Also, results will be presented at national and international cardiovascular conferences. The Chief Investigator (CI) of this study is in an ideal position to coordinate the dissemination of the results of this study as they are an active member of several research consortia and professional organisations. Among others, he chairs the board of the German Atrial Fibrillation competence NETwork (AFNET) and the Education Committee of the European Society of Cardiology. The CI sits on the Board of the European Heart Rhythm Association and the European Society of Cardiology. The CI is Deputy Editor of Heart, an Editorial Board Member of BMC Medicine, Heart Rhythm, and Europace, and serves as Editorial Consultant to The Lancet. All original research submitted for publication to international peer-reviewed scientific journals will be subject to copyright legislation. Data collection will be completed following the acquisition of mortality data from NHS Digital. Following a period of analysis and write up, the study team aim to publish results from this study in September 2020. Patients with AF who are treated at Sandwell and West Birmingham Hospitals NHS Trust are regularly invited to attend AF meetings designed to update patients on new developments in the diagnosis and treatment of AF and relevant developments in local AF services. The BBC-AF study was a local cohort study which recruited a large number of patients with AF who use local AF services. Many patients who participated in this study are likely to be very interested in the findings of this study and to discover how the results of this study could improve the care delivered to patients with AF. As such, on completion of this study, a patient meeting will be organised which will be open to all local patients who have an interest in AF in order to present the output from the BBC-AF study. This meeting will be free to attend and will aim to discuss and communicate study findings and their potential significance for patient care.

Processing:

The proposed study will determine if any of the participants recruited to the BBC-AF study are alive or dead. This application aims to determine the date of death and cause of death from the NHS Digital's Civil Registration (Deaths) extract and Demographics extract. To do this, the study team will provide a cohort containing the Study ID, and patient identifiers [NHS Number, Date of Birth and Postcode] for each of the cohort members included in this study. COHORT SIZE A total of 1,630 patients, referred to the SWBH NHS Trust, for inpatient or outpatient evaluation of acute illnesses were recruited between September 2014 and February 2018 for the Birmingham and Black Country Atrial Fibrillation Registry (BBC-AF). The start date for this study and the date the first patient was recruited to this study was 23 April 2014. NHS Digital has found that for the 143 study participants recruited using the Patient Consent materials (v4) between September 2014 to June 2015 consent was not compatible for this data request, and therefore those initial 143 participants have been excluded from this application. For the purpose of this application, the cohort size will be 1,487 consisting only of those participants recruited under v7 consent material between June 2015 to February 2018. The cohort is a fixed size. METHOD: - SWBH NHS Trust provide the Study ID plus identifiers [NHS Number, Date of Birth and Postcode] from a cohort of 1,487 individuals to the Cohort Management System (CMS) via SEFT. - CMS automatically produces the Civil Registration (Deaths) extract and Demographics extract. - CMS supplies the pseudonymised reports back to SWBH NHS Trust via SEFT. - SWBH NHS Trust links the pseudonymised Cause of Death report from NHS Digital to the pseudonymised BBC-AF study data base. - SWBH NHS Trust will therefore ensure that all data is pseudonymised before it is made available to staff at the University of Birmingham. NHS Digital data will not be linked with any identifying data once it has been disseminated by NHS Digital. Only pseudonymised data will be available to the University of Birmingham for analysis using the Castor Platform. Provision of the study numbers will facilitate the return of data containing study ID only from NHS Digital thereby further restricting the level of confidential data being transferred from NHS Digital to SWBH NHS Trust. STORAGE All patient data obtained from NHS Digital will be stored at the Clinical Research Facility at Sandwell General Hospital. Data will be stored in line NHS Code of Practice relating to Confidentiality. Confidential data containing patient identifiers will only be stored at SWBH NHS Trust. The master recruitment log for this study is stored securely at the clinical research facility at Sandwell Hospital. All patients in this study have a unique study number. Information on patients in this study is stored on a pseudonymised electronic case report form (Castor Platform). Castor is a web-based system that enables researchers to build electronic Case Report Forms (eCRFs) that enables safe and valid data collection. All patients recruited to the BBC AF study already have an eCRF on Castor. Each patient has a unique study ID which can identify each patient from the master recruitment log which is stored at the Clinical Research Facility at Sandwell General Hospital. Pseudonymised mortality data obtained from NHS Digital will therefore only be used to update the eCRF for those patients who have died. The mortality information will be uploaded onto the Castor platform. Thereafter, data analysis will be done using the pseudonymised data base. Each participant in this study already has an electronic case report form on the Castor Platform. This electronic case report form is pseudonymised with each participant having a study ID but no patient identifiers. The follow-up section of this form contains information on each participant's current status i.e. alive or dead. For those participant who are deceased, the case report form also has a section on the cause of death to determine if this was related to cardiovascular disease. All pseudonymised data obtained from NHS Digital will therefore be used to update each of the case report forms by using the study ID to match this data. The study team at the University of Birmingham will not be able to see identifying patient information at any time. The research lead working on this study is a Cardiology Registrar currently working as a research fellow at SWBH NHS Trust. The University of Birmingham has an Information Security Management System based on ISO27001 with a range of controls covering the protection of personal information. Annual security awareness training is mandatory for staff and the University is accredited under the NHS Information Governance Toolkit. The University’s Information Security Policy is a policy as defined in University regulations and its observance is mandatory for all computer users and for all computer equipment and electronic devices that connect with the University’s IT network. Only two researcher from the University of Birmingham have password access to pseudonymised ​ data on the Castor platform. As per University of Birmingham data protection guidelines, this data can only be accessed when researchers are using the University's secure password protected IT network. This data can only be stored and accessible from the University of Birminghams IT network. Access to this data on an unsecure remote devices or unsecure internet networks would be in breach of University of Birmingham IT policy and therefore is not permitted. All data relating to participants recruited to this study have been pseudonymised. That is to say, each participant has a unique study ID that can only be used to identify each participant by using the master recruitment log which is stored at the Clinical Research Facility at Sandwell General Hospital. Data obtained from NHS Digital will also be pseudonymised by also using the study ID. In addition, this data will be stored at SWBH NHS Trust for the minimum period required for the analysis of this data i.e. 2 years only. All pseudonymised data obtained from NHS Digital will be uploaded onto the pseudonymised e-CRF. There will be no attempt to re-identify individuals from this study or linkage to other data sets other than as set out in this agreement. All patient data is encrypted and only substantive employees of SWBH NHS Trust can access confidential patient data. At present, data relating to clinical studies is stored on a server "S-drive". Access to this drive is password protected and only substantive employees of SWBH NHS Trust who have been granted access by their line managers can access these files. Therefore, only staff based at SWBH NHS Trust who are actively involved in research will have permission to access these files. Also, access can only be achieved from an NHS encrypted computer. All data storage is in line with the NHS Confidentiality Code of Practice. The Castor servers are located in the United Kingdom and are ISO 27001 and ISO 9001 certified. Encryption is applied to all sensitive data stored in Castor. Castor is fully compliant with GDPR guidelines. The Castor system uses Microsoft Azure as a back-up storage facility. The data will be stored in Microsoft's London data centre (UK South), and backups will be stored in the Cardiff (UK West) data centre. Microsoft Ltd provide Azure Backup Storage Services for Sandwell and West Birmingham Hospitals NHS Trust and are therefore listed as a data processor. They supply support to the system, but do not access data. Therefore, any access to the data held under this agreement would be considered a breach of the agreement. This includes granting of access to the database[s] containing the data.