| FULL TITLE South London Registry of Cardiovascular Diseases SHORT TITLE SoLoR-CVD | ||
| Sponsor: | King’s College Hospital NHS Foundation Trust | |
| Funder (s): | King’s College Hospital Cardiovascular Research Fund | |
| IRAS Reference | 321402 | |
| ISRCTN / Clinicaltrials.gov no: | N/A | |
Protocol Version and Date v2.0 21/July/2025
KEY ROLES AND RESPONSIBILITIES
SPONSOR: The sponsor is responsible for ensuring before a research project begins that arrangements are in place for the research team to access resources and support to deliver the research as proposed and allocate responsibilities for the management, monitoring and reporting of the research. The Sponsor also has to be satisfied there is agreement on appropriate arrangements to record, report and review significant developments as the research proceeds, and approve any modifications to the design.
FUNDER: The funder is the entity that will provide the funds (financial support) for the conduction of the project. Funders are expected to provide assistance to any enquiry, audit or investigation related to the funded work.
CHIEF INVESTIGATOR (CI): The person who takes overall responsibility for the design, conduct and reporting of a project. If the research involves researchers at more than once site, the CI takes on the primary responsibility whether or not he/she is an investigator at any particular site.
The CI role is to complete and to ensure that all relevant regulatory approvals are in place before the project begins. Ensure arrangements are in place for good conduct, robust monitoring and reporting, including prompt reporting of incidents, this includes putting in place adequate training for research staff to conduct the project as per the protocol and relevant standards.
The Chief Investigator is responsible for submission of annual reports as required. The Chief Investigator will notify the R&I Office of the end of the project, including the reasons for the premature termination. Within one year after the end of project, the Chief Investigator will submit a final report with the results, including any publications/abstracts to the REC.
PRINCIPAL INVESTIGATOR (PI): Individually or as leader of the researchers at a site; ensuring that the project is conducted as per the approved registry protocol, and report/notify the relevant parties – this includes the CI of any breaches or incidents related to the registry.
DECLARATIONS
The undersigned confirm that the following protocol has been agreed and accepted and that the investigator agrees to conduct the research in compliance with the approved protocol and will adhere to the Research Governance Framework 2005 (as amended thereafter), the Trust Data & Information Governance policy, Sponsor and other relevant SOPs and applicable Trust policies and legal frameworks.
I (investigator) agree to ensure that the confidential information contained in this document will not be used for any other purposes other than the evaluation or conduct of this research without the prior written consent of the Sponsor.
I (investigator) also confirm that an honest accurate and transparent account of the research will be given; and that any deviations from the research as planned in this protocol will be explained and reported accordingly.
Chief Investigator:
Signature: Date: 21/July/2025
Print Name (in full): Dr Nilesh Pareek
Position: Adjunct Senior Lecturer, King’s College London
Consultant Interventional Cardiologist, King’s College Hospital
CONTENTS
11 DATA HANDLING AND MANAGEMENT. 12
12 MATERIAL/SAMPLE STORAGE. 12
13 PEER AND REGULATORY REVIEW… 12
14 ASSESMENT AND MANAGEMENT OF RISK. 12
15 GENERAL DATA PROTECTION REGULATION.. 12
16 MONITORING AND AUDITING.. 13
21 PUBLICATION AND DISSEMINATION POLICY. 14
KEY WORDS
Acute coronary syndrome
Atrial Fibrillation
Aortic Stenosis
Electronic health record
Mitral regurgitation
Natural language processing
Ventricular tachycardia
Heart failure
LIST OF ABBREVIATIONS
| ACS | Acute Coronary Syndrome |
| AE | Adverse Event |
| AF | Atrial Fibrillation |
| AR | Adverse Reaction |
| AS | Aortic Stenosis |
| CAG | Confidential Advisory Group |
| CCS | Chronic Coronary Syndrome |
| CI | Chief Investigator |
| CVD | Cardiovascular disease |
| DMC | Data Monitoring Committee |
| EHR | Electronic Health Record |
| EP GStT | Electrophysiology Guy’s and St Thomas’ Hospital |
| HRA | Health Research Authority |
| ISRCTN KCH NHS | International Standard Randomised Controlled Studies Number King’s College Hospital National Health Service |
| NLP PCI | Natural Language Processing Percutaneous Coronary Intervention |
| PI | Principal Investigator |
| QA | Quality Assurance |
| QC | Quality Control |
| PCI | Percutaneous Coronary Intervention |
| REC | Research Ethics committee |
| SAR | Serious Adverse Reaction |
| SAE | Serious Adverse Event |
| SDV | Source Data Verification |
| SOP | Standard Operating Procedure |
| STEMI | ST-segment elevation myocardial infarction |
| VT | Ventricular Tachycardia |
Trial personnel
Chief Investigator Dr Nilesh Pareek
Department of Cardiology, King’s College Hospital, Denmark Hill, London, SE5 9RS
tel: +44 (0)20 3299 9000
Co-Investigators Dr Rafal Dworakowski
Department of Cardiology, King’s College Hospital, Denmark Hill, London, SE5 9RS
tel: +44 (0)20 3299 9000
Dr Michael McGarvey
Department of Cardiology, King’s College Hospital, Denmark Hill, London, SE5 9RS
tel: +44 (0)20 3299 9000
1 SUMMARY
| OVERVIEW | |
| Full title | South London Registry of Cardiovascular Diseases |
| Objectives | Primary objective: To describe the natural history of cardiovascular diseases in a contemporary, urban UK population Secondary objective: To validate existing treatment algorithms and risk-prediction tools against patient-oriented outcomes in a large, contemporary real-world cohort To derive novel risk prediction tools that integrate structured and unstructured data inputs, to improve risk stratification and inform the development of personalised treatment plans using data collected during routine delivery of care |
| Type of trial | Retrospective disease registry |
| Trial design and methods | This is a two-centre disease registry from two large, academic cardiac centres (King’s College Hospital and Guy’s & St Thomas’ Hospital) within the United Kingdom. Participants: Participants are those who have been inpatients or outpatients treated for cardiovascular disease, at either hospital, between 2012-2022. |
| Health condition(s) or problem(s) studied | Cardiovascular conditions including acute and chronic coronary syndromes, structural and valvular heart disease, arrhythmia and complex electrophysiology and, heart failure and cardiomyopathy |
| Target sample size | N/A |
| Trial duration per participant: | N/A |
| Main inclusion/exclusion criteria: | Inclusion criteria (all): Age ≥ 18Documented history of a cardiovascular conditionEither: a) a hospital episode in cardiology out-patients, or b) an in-patient episode with a cardiovascular condition stated as the primary or secondary diagnosis, between 2012 – 2022 Exclusion criteria (any): Previously expressed wish to opt-out of healthcare research or be excluded from the SoLoR-CVD |
| Statistical methodology and analysis: | We anticipate that the SoLoR-CVD registry will provide a rich research resource and provide the basis for a series of manuscripts. Each separate analysis will have a dedicated statistical plan. Continuous data will be summarised as mean and standard deviation or median with interquartile range. Continuous variables will be compared using Student t test or, if not normally distributed, using Mann-Whitney U test. Categorical data will be expressed as absolute frequencies and percentages and compared with X2 tests. Patients may be grouped by age category, gender or ethnicity depending on the analysis. For the derivation of novel risk prediction tools, selected candidate variables will undergo univariable analysis. Following this, a smaller subset of predictor variables will be entered into a multivariable analysis. A receiver operating characteristic (ROC) curve and area under the curve analysis will be performed to assess performance. Time to event analysis will be used to assess clinical endpoints and expressed as Kaplan-Meier estimates. |
| TIMELINES | |
| Duration/length | 10 years |
| Expected Start Date | August 2023 |
| End of Project definition and anticipated date | This will be a retrospective disease registry. The database will be stored for 10 years. |
| Data collected / Storage | Identifiable personal data will be extracted from the electronic patient record and be de-identified before being stored on NHS computers. A unique identification number will be assigned upon entering the registry. |
2 INTRODUCTION
Cardiovascular disease (CVD) is the leading cause of adult death and morbidity worldwide, affecting up to 7 million people in England alone, causing one in four of premature adult deaths. CVD contributes significantly to health inequalities and disproportionately affects those with lowest incomes, and people from South Asian and Black communities (1). Improving the prevention and treatment of cardiovascular disease and reducing resultant health-related social inequalities, has been identified as a key priority for the National Health Service (2).
Harnessing the power of data collected during routine delivery of care is central to the NHS strategy for improving outcomes in CVD (3). Prospective cohort studies such as the Framingham Heart Study (4) and the PURE study (5) have provided important insights into our understanding of the pathophysiology, risk factors and the natural history of CVD. Whilst valuable, such studies are time-consuming and resource intensive.
Advances in bioinformatics enable large datasets to be collected, processed, and rapidly analysed. This presents a unique opportunity given the volume of high-quality data collected during routine care within the electronic health record (EHR). This high-quality data can be linked and analysed in concert with external datasets to gain novel and real-time insight into the evolving field of CVD.
3 BACKGROUND AND RATIONALE
For over 10 years, King’s College Hospital NHS Foundation Trust (KCH) and Guy’s & St Thomas’ NHS Foundation Trust (GStT) have operated a fully electronic patient health record (EHR.) The EHR represents longitudinal data collected during the routine delivery of care, containing a depth of data on patient demographics, symptoms, co-morbidities, medication use and treatment outcomes that is unsurpassed in prospectively gathered datasets (6). This rich dataset can be linked to data collected in other sectors (e.g., education, housing, social care, criminal justice) to support a range of research projects on the natural history of disease, drug utilisation, safety and efficacy of specific treatment interventions and the societal impact of cardiovascular diseases in large, real-world populations (7).
The SLAM Biomedical Research Centre (BRC) Case register was founded in 2007 and represents a class-leading example of how routinely collected EHR data can be used in research. Using novel, user friendly tools for data-extraction, the SLAM-BRC Case register allows rich and detailed patient phenotyping and has provided key insights in the fields of psychiatry and psychological medicine (8). In cardiovascular medicine, national disease registries such as the Myocardial Ischaemia National Audit Project (MINAP) (9) and ‘big data’ projects such as the NIHR Health Informatics Collaborative (10, 11) provide important insights, but lack the depth of data required to permit direct application to routine practice.
The South London Registry of Cardiovascular Diseases (SoLoR-CVD), will provide a resource that is rich both in its breadth and depth of data, spanning ischaemic heart disease, valvular heart disease, complex arrhythmias and heart failure. Using the Cogstack data-extraction application ecosystem, data will be extracted from structured (e.g., demographics, vital signs, laboratory test results) and unstructured (e.g., clinical notes, procedure reports) data-sources within the EHR (12). Unstructured data from ‘free-text’ sources will undergo natural language processing (NLP) into analysable format using the self-supervised, machine learning Medical Concept Annotation Toolkit (MedCAT) (13). Approximately 80% of data within the EHR is stored within unstructured text – NLP represents a powerful but underutilised method for unlocking this information and enriching our understanding of cardiovascular disease in contemporary, real-world populations (14).
The SoLoR-CVD will be a rich and unique contemporary dataset, from a large and diverse, urban population with cardiovascular disease. The Registry will employ field-leading techniques for data-extraction, using artificial intelligence, powered by machine-learning algorithms to create an invaluable resource for clinical researchers, care deliverers and policy researchers in the field of cardiovascular medicine.
4 OBJECTIVES
4.1 Primary Objective
4.2 Secondary Objectives
- To validate existing treatment algorithms and risk-prediction tools against patient-oriented outcomes in a large, contemporary real-world cohort
- To derive novel risk prediction tools that integrate structured and unstructured data inputs, to improve risk stratification and inform the development of personalised treatment plans using data collected during routine delivery of care
5 DESIGN
Location:
This will be a two-centre, retrospective disease registry, based at two large tertiary cardiac referral centres in South London: Guy’s & St Thomas’ NHS Foundation Trust & King’s College Hospital NHS Foundation Trust.
Participants:
The Registry will include all adult patients, with a history cardiovascular disease, who have had a cardiology outpatient episode at KCH or GStT, or an in-patient episode with a cardiovascular condition as a primary or secondary diagnosis, between April 2012 and March 2022.
Protocol:
The SoLoR-CVD is a de-identified database derived from the electronic health records (EHR) of two large university hospitals, that provide highly specialised cardiac care to a socially diverse multi-ethnic population in London and the south-east of England. The services include a comprehensive range of cardiac sub-specialties such as percutaneous coronary interventions (PCI) for acute and chronic coronary syndromes, percutaneous heart valve and structural interventions, complex electrophysiology and arrhythmia management, advanced cardiovascular imaging and heart failure services. Since 2010, both hospitals have operated a fully electronic health record (EHR).
DATA EXTRACTION & DE-IDENTIFICATION
A locally developed information retrieval system, Cogstack, will be used to extract information from the EHR at each hospital into a site-specific research database, namely: i) King’s Electronic Records Research Interface (KERRI), and ii) Guy’s & St Thomas’ Electronic Records Research Interface (GERRI.)
During extraction, identifiable data will be removed (name, hospital ID and NHS ID) or weakened by truncation (i.e., month and year of birth; postcode modified to Lower Layer Super Output Area; ethnic category collapsed into the NHS standard 16+1 categories). Eligible participants will be assigned a Registry specific ID. A linkage file will be maintained separately from the clinical data and will link Registry specific ID to NHS ID and date of birth.
The initial data extraction occurs independently at each site. Following extraction and creation of a site-specific clinical registry and linkage file, the GStT data files will be transferred to KCH via a secure, NHS encrypted connection. Due to the shared geographical catchment area, there is risk of duplication within the registry. The GStT and KCH Linkage files will be combined into a SoLoR-CVD linkage file to facilitate de-duplication and ensure that a single Registry ID is issued per NHS ID. The linkage file will only be accessible to the CI and database manager. The GStT clinical registry and the KCH clinical registry will be combined to create the South London Registry of Cardiovascular Diseases and will be accessible to all members of the database research team.
Cardiac and intracoronary imaging is integral to the assessment and understanding of cardiovascular disease and its clinical manifestations. Ultrasound (intravascular ultrasound (IVUS) and echocardiography), computerised tomography (CT), positron emission tomography (PET), intracoronary optical coherence tomography (OCT), and cardiac magnetic resonance imaging (CMR) modalities are applied in daily practice. The raw-imaging data remains either on a dedicated ‘offline review system’ or a digital picture archiving and communication system (PACS) within encrypted and protected NHS data infrastructure. Raw-imaging data is processed in dedicated software, purchased by KCH & GStT, to facilitate routine care. This data can be viewed, analysed, and reported by clinicians to assist with patient diagnosis and assessment. The generated reports are integrated into the EHR and will be extracted into the KERRI and GERRI databases, but the raw imaging data remains separate.
SoLoR-CVD researchers may wish to apply extended analyses of the raw imaging data for the purpose of research. Analyses, either manual or automated, can be performed on dedicated NHS image review stations within the KCH/GStT NHS data-security infrastructure, using vendor supplied software purchased by KCH/GStT. In some cases, automated image analysis may only be possible using a third-party software (e.g., provided by a university or industry collaborator.) Third parties would be permitted to access to de-identified raw-data files (labelled by registry-specific ID) in an NHS data safe-haven or through transfer of images via an approved, secure connection. Once the analysis was complete, third-parties would be required to delete the raw-data file. The output of any image analysis (manual or automated) can be entered or extracted into dedicated databases that can be linked with the SoLoR-CVD database, using appropriate linkage files. Linkage files, including date of birth, NHS ID and Registry ID would remain within the Trust firewall.
To ensure accurate capture of important patient-oriented outcomes including hospital readmission, cause of re-admission and cause of death, the SoLoR-CVD database will be linked to external data sources i.e., NHS Hospital Episode Statistics and the Civil Registration of Deaths. KCH will provide a trusted third party (NHS England) with the SoLoR-CVD Linkage File. NHS England will perform the linkage and return the linked dataset to KCH.
Data extraction from the EHR and linkage with external datasets will be performed at a single timepoint to cover the full cohort and ensure a minimum of 1-year clinical follow-up.
RESEARCH & QUALITY IMPROVEMENT
The SoLoR-CVD database will support high-impact research into cardiovascular disorders and their care across four principal domains:
a) Acute and Chronic Coronary Syndromes;
b) Structural & Valvular Heart Disease;
c) Arrhythmia & Complex Electrophysiology; and
d) Heart Failure & Cardiomyopathy.
The registry will also provide a rich resource to improve upon the quality of routinely delivered care.
Researchers or clinicians wishing to access the SoLoR-CVD database will be required to have a substantive contract with KCH, or a substantive contract with GStT or KCL and honorary contract with KCH. To gain access to the registry, they will be required to submit a written application to the Oversight Committee outlining their proposal and required dataset. The Oversight Committee will review the proposal in respect to the scientific validity and clinical importance, with particular attention paid to potential commercial implications and the risk for patient re-identification (e.g., projects with low patient numbers with high risk of re-identification will be discouraged.) If an approval is issued, the researchers will receive the minimal required dataset for their analysis.
Approved projects may involve collaboration with other research groups within the NHS and associated universities or other third parties (e.g., non-NHS researchers or industry collaborators.) External collaborators may provide independent verification of methods of data-analysis and interpretation, or validate novel methods of analysis (e.g., in the interpretation of cardiac or coronary imaging.) Such collaborations may involve the sharing of de-identified data sets, or de-identified raw-imaging data (labelled only with registry specific ID). In all cases, the minimal required de-identified dataset would be shared via an NHS data safe-haven or via secure servers – collaborators would be required to delete any shared dataset once the analysis was complete. Applications where commercial partners stand to gain disproportionately from data access, relative to potential benefit to health and social care system, may not be approved.
It is envisaged that the SoLoR-CVD database will support the publication of numerous peer-reviewed scientific publications assessing the natural history of cardiovascular disease, description of novel risk factors and assessment of specific treatment strategies in a large, contemporary, socially diverse, and multi-ethnic cohort. External collaboration and validation of novel methods and will ensure that this research is of high scientific quality, with maximal impact from resultant publications.
6 CONSENT
The SoLoR-CVD database will include hundreds of thousands of individual patients, spanning 10 years of clinical encounters. It is not felt that it would be practical or feasible to seek individual consent for participation. At the point of entry to the Registry, patient identifiable data will be removed (e.g., name, hospital ID, NHS ID), truncated or weakened (e.g., post-code, date of birth, procedure dates.) Each patient will be issued with a unique, registry specific identification code linked to the source system ID and NHS ID via dedicated linkage tables. The CI and senior database administrator will be the only members of the SoLoR-CVD team who will have access to the linkage files.
The decision to follow an approach of ‘consent or anonymise’ follows UK case law interpretation of the European Data Directive (the ‘Source Informatics’ ruling) which establishes that the use of data for purposes other than that for which it was created does not require consent by the providers of the data, provided that the data are anonymous. This approach remains in accordance with UK Data Protection Act (2018). The same measures will be applied to people who are deceased, in accordance with common law duty of confidentiality.
Extensive efforts will be made to ensure that the SoLoR-CVD is widely publicised to raise patient awareness. An ‘opt out’ facility will be advertised, in addition to clear instructions for patients to express their wish to ‘opt out’. Patients may hay have previously registered with the National Data Opt-out scheme – this choice is submitted by local GPs and registered with NHS Digital. All relevant patients will be removed before entering the Registry.
Patient and public involvement (PPI) will be central throughout the design and management of the SoLoR-CVD. A Patient and Public Oversight Group consisting of five people with diverse backgrounds (including, those with experience of cardiovascular disease or as a carer for someone with cardiovascular disease) has been created to provide guidance and support to the Registry team. They provide an intermediary between researchers and our patient body and represent a resource that researchers can draw upon to ensure that studies reflect the needs of our local population, to prepare and inform grant applications, guide research conduct and disseminate results in an appropriate and accessible format. During the ‘initiation phase’, the PPI Oversight Group have met on an ad-hoc basis. Once the Registry is established, they will meet on a 6-monthly basis. In addition to this group, we will seek ongoing input from our local population through regular campaigns to engage our local population, and seek their opinions to guide the evolution of the project.
The Registry Oversight Committee will include the CI, who will act as the data-custodian, in addition to senior researchers and cardiovascular physicians. Researchers and clinicians wishing to access the SoLoR-CVD will have to apply to the Oversight Committee with a summary of their planned research and the dataset they require. The oversight Committee will review the proposal in respect to the scientific validity, potential commercial implications and the risk for potential patient re-identification. If approval is issued, the researchers will receive the minimal required dataset for their analysis.
7 ELIGIBILITY CRITERIA
7.1 Inclusion Criteria
- A documented cardiovascular condition including: acute and chronic coronary syndromes; valvular heart disease; arrhythmia; and heart failure
- Documented in-patient episode with cardiovascular conditions stated as a primary or secondary diagnosis from April 2012 – March 2022 and/or documented out-patient episode in a cardiology clinic
- Age >18 yrs
7.2 Exclusion Criteria
- Previously expressed wish to opt-out of healthcare research via the NHS National data optout scheme or an expressed wish be excluded from the SoLoR-CVD
8 RECRUITMENT
Participants will not be actively recruited. This is a retrospective registry, and patients will be included if they are over the age of 18, have a documented history of a cardiovascular condition and have had either: i) a hospital episode in cardiology out-patients, or ii) an in-patient episode with a cardiovascular condition stated as the primary or secondary diagnosis.
9 STATISTICAL METHODS
The SoLoR-CVD will be a large, longitudinal disease registry that will support production of a series of peer-reviewed scientific publications. Each manuscript will have a statistical plan specific to its stated research question.
Broadly, continuous data will be summarised as mean and standard deviation or median with interquartile range. Continuous variables will be compared using Student t test or, if not normally distributed, using Mann-Whitney U test. Categorical data will be expressed as absolute frequencies and percentages and compared with X2 tests. Patients may be grouped by age category, gender or ethnicity depending on the analysis.
For the derivation of novel risk prediction tools, selected candidate variables will undergo univariable analysis. Following this, a smaller subset of predictor variables will be entered into a multivariable analysis. A receiver operating characteristic (ROC) curve and area under the curve analysis will be performed to assess performance. Time to event analysis will be used to assess clinical endpoints and expressed as Kaplan-Meier estimates. Statistical analysis will be performed using R Software Version 4.2.1.
10 FUNDING
The SoLoR-CVD will be funded using locally supplied research grants from the King’s College Hospital Cardiovascular Department, including support from the William Frederick Haynes Trust. As the Registry develops, grant applications will be made to external bodies to support the development and evolution of the project.
11 DATA HANDLING AND MANAGEMENT
The SoLoR-CVD will be managed from the King’s College Cardiovascular Research Unit, based at King’s College Hospital, Denmark Hill. Staffing for the registry will include:
1) The Chief Investigator (10% WTE) – academic lead; shared leadership of the data resource
2) Senior database administrator (10% WTE)
3) Clinical Informatics lead (10% WTE) – shared leadership of the data resource
4) Four Cardiovascular Theme Leads (5% WTE / person)
5) PhD student / pre-doctoral researcher (100% WTE)
The Chief Investigator will act as custodian for the Registry data. The following guidelines will be strictly adhered to:
- Patient data will be de-identified at the point of data extraction
- Each patient will be given a unique registry identification number
- All de-identified data will be stored on a password protected computer within an NHS Trust Firewall
- All data sharing parties will be required to have a substantive contract with KCH (or a substantive contract with GStT/KCL and honorary contract with KCH), will have completed NIHR Good Clinical Practice (GCP) or equivalent training, and will meet current data-handling and clinical governance standards and guidelines.
- All projects that require Registry data will require approval from an Oversight Committee – applications will be reviewed in respect to the scientific validity, clinical value, any commercial implications and the risk for potential patient re-identification. If approval is issued, the researchers will receive the minimal required dataset for their analysis.
- All activity related to access of the Registry is logged in an audit trail, accessible to the Registry administrators only
- Patients have the right to opt-out from having their record the SoLoR-CVD, including those who have previously expressed their wish (via their GP) to be added to the National Data Opt-out list.
- Data-linkage to external datasets will be carried out using minimal required data (e.g., NHS number and data of birth) and trusted third parties i.e., NHS England. Once data-linkage had been completed, the Third-party will destroy patient-identifiable data.
- In order to perform or validate novel methods of image or data analysis, de-identified data may be shared with third-party collaborators (e.g., university statisticians, other NHS researchers, or industry collaborators) provided the following criteria are satisfied:
(a) the minimal required dataset is shared;
(b) all shared data is de-identified;
(c) data is shared through an approved and secure connection;
(d) collaborators must destroy all data once validation analysis is complete;
(e) a ‘Data Protection Impact Assessment’ has been completed and approved by the Sponsor Information Governance Office (IGO)
(f) a ‘Data Sharing Agreement’, signed off by the Sponsor Caldicott Guardian, is mandatory for all industry and non-UK collaborations, or when requested by the IGO following completion of the Data Protection Impact Assessment.
12 MATERIAL/SAMPLE STORAGE
Due to the nature of the disease registry, no biological material will be gathered from patients.
13 PEER AND REGULATORY REVIEW
The registry has been peer reviewed in accordance with the requirements outlined by KCH R&I. The registry was has received regulatory approval from Health Research Authority (HRA)(IRAS 321402), Research Ethics Committee (REC) (23/NI/0077) & Clinical Advisory Group (CAG) (23/CAG/0090).
14 ASSESSMENT AND MANAGEMENT OF RISK
The retrospective and observational nature of the Registry mean that risks to individuals will be minimal and limited to the risk for patient re-identification.
Extensive precautions will be taken to minimise the risk of re-identification. The Registry will be subject to regular and ongoing audit. Processes will be reviewed at quarterly meetings of the Registry Oversight Committee. The Oversight Committee is accountable to the Sponsor, the Trust Caldicott Committee and Trust Information Governance representatives.
15 GENERAL DATA PROTECTION REGULATION
King’s College Hospital is the sponsor for this Registry, which is based at 2 large academic cardiology centres in London, United Kingdom. Data extracted for purposes of the Registry will be handled and stored in accordance with the Data Protection Act, 2018.
The Chief Investigator will act as custodian for the Registry trial data and the guidelines outlined above (Section 11) will be strictly adhered to.
All data will be de-identified and stored in a dedicated Registry database. Researchers or clinicians (not listed within the named Registry staff) will have to submit a project-specific application to access Registry data. If access is granted, they will be given the minimal required dataset.
All data will be maintained by King’s College Hospital NHS Foundation Trust and the Registry administrator. Linkage to external datasets, using trusted third parties (i.e., NHS England) will be performed to ensure a minimum of 1 year clinical follow-up.
All data-sharing parties will be required to have substantive or honorary contracts with KCH, will have completed GCP or equivalent training, and will be subject to current data-handling and clinical governance standards and guidelines.
16 MONITORING AND AUDITING
The Chief Investigator will ensure that there are adequate quality and number of monitoring activities conducted by the Registry management team and sponsor. This will include adherence to the protocol and procedures to ensure adequate data quality and governance.
The Chief Investigator will inform the sponsor should they have concerns which have arisen from monitoring activities, and/or if there are problems with oversight or monitoring procedures.
17 TRAINING
The Chief Investigator will review and provide assurances of the training and experience of all staff working on this project. Appropriate training records will be maintained in the Registry files.
18 INDEMNITY ARRANGEMENTS
KCH will provide NHS indemnity cover for negligent harm as appropriate and is not in the position to indemnify for non-negligent harm. NHS indemnity arrangements do not extend to non-negligent harm and NHS bodies cannot purchase commercial insurance for this purpose; it cannot give advance undertaking to pay compensation when there is no negligence attributable to their vicarious liability. The Trust will only extend NHS indemnity cover for negligent harm to its employees, both substantive and honorary, conducting research studies that have been approved by the R&D Department. The Trust cannot accept liability for any activity that has not been properly registered, and Trust approved. Potential claims should be reported immediately to the Joint Research Office.
19 ARCHIVING
The SoLoR-CVD will be a retrospective disease registry. Data will be fully de-identified during extraction and stored on secure servers accessible only to the research team/co-investigators. Linkage to external datasets, using trusted third parties (i.e., NHS England) will be performed to ensure a minimum of 12 months clinical follow-up. Data will be stored for 10 years after the completion of the data linkage.
20 PUBLICATION AND DISSEMINATION POLICY
The SoLoR-CVD will support the production of a series of peer-reviewed scientific manuscripts. Results will be disseminated via peer-reviewed scientific journals and presentations at National and International Conferences. No identifiable patient data will be used during dissemination of results.
Registry participants will not be directly informed of results, but extensive efforts will be made to work collaboratively with partners in PPI to disseminate the results of research generated from the Registry. The PPI Oversight Committee will meet on an annual basis to review Registry processes. They will advise on the production of plain-English summaries of any published manuscripts. The plain-English summaries will be disseminated in poster format (for placement in hospital waiting rooms, clinic rooms etc), and via websites for associated partners (e.g., KCH & GStT websites; KCL BHF Centre of Excellence website) and via their associated social media profiles.
REFERENCES
1. England PH. Health matters: preventing cardiovascular disease Gov.UK Website2019 [Available from: https://www.gov.uk/government/publications/health-matters-preventing-cardiovascular-disease/health-matters-preventing-cardiovascular-disease.
2. Care DoHS. The NHS Long Term PLan. 2019.
3. Care DoHS. Better, broader, safer: using health data for research and analysis. Gov.uk website; 2022.
4. Sytkowski PA, Kannel WB, D’Agostino RB. Changes in Risk Factors and the Decline in Mortality from Cardiovascular Disease. New England Journal of Medicine. 1990;322(23):1635-41.
5. Yusuf S, Joseph P, Rangarajan S, Islam S, Mente A, Hystad P, et al. Modifiable risk factors, cardiovascular disease, and mortality in 155 722 individuals from 21 high-income, middle-income, and low-income countries (PURE): a prospective cohort study. The Lancet. 2020;395(10226):795-808.
6. Denaxas SC, Morley KI. Big biomedical data and cardiovascular disease research: opportunities and challenges. European Heart Journal – Quality of Care and Clinical Outcomes. 2015;1(1):9-16.
7. Cowie MR, Blomster JI, Curtis LH, Duclaux S, Ford I, Fritz F, et al. Electronic health records to facilitate clinical research. Clinical Research in Cardiology. 2017;106(1):1-9.
8. Perera G, Broadbent M, Callard F, Chang C-K, Downs J, Dutta R, et al. Cohort profile of the South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLaM BRC) Case Register: current status and recent enhancement of an Electronic Mental Health Record-derived data resource. BMJ Open. 2016;6(3):e008721.
9. Herrett E, Smeeth L, Walker L, Weston C. The Myocardial Ischaemia National Audit Project (MINAP). Heart. 2010;96(16):1264-7.
10. Kaura A, Panoulas V, Glampson B, Davies J, Mulla A, Woods K, et al. Association of troponin level and age with mortality in 250 000 patients: cohort study across five UK acute care centres. BMJ. 2019:l6055.
11. Kaura A, Sterne JAC, Trickey A, Abbott S, Mulla A, Glampson B, et al. Invasive versus non-invasive management of older patients with non-ST elevation myocardial infarction (SENIOR-NSTEMI): a cohort study based on routine clinical data. The Lancet. 2020;396(10251):623-34.
12. Jackson R, Kartoglu I, Stringer C, Gorrell G, Roberts A, Song X, et al. CogStack – experiences of deploying integrated information retrieval and extraction services in a large National Health Service Foundation Trust hospital. BMC Medical Informatics and Decision Making. 2018;18(1).
13. Kraljevic Z, Searle T, Shek A, Roguski L, Noor K, Bean D, et al. Multi-domain clinical natural language processing with MedCAT: The Medical Concept Annotation Toolkit. Artificial Intelligence in Medicine. 2021;117:102083.
14. Reading Turchioe M, Volodarskiy A, Pathak J, Wright DN, Tcheng JE, Slotwiner D. Systematic review of current natural language processing methods and applications in cardiology. Heart. 2022;108(12):909-16.
SoLoR-CVD 