Competing interests None declared.
Provenance and peer review Commissioned; internally peer reviewed.
Statistics from Altmetric.com
- Transcatheter aortic valve implantation
- coronary artery disease
- coronary intervention
- interventional cardiology
- percutaneous coronary intervention
- interventional cardiology
- non-coronary intervention
The Transcatheter Aortic Valve Implantation (TAVI) Steering Group was established in 2008 to guide the introduction of this novel therapy which had been undertaken for the first time in the UK in 2007. The group was set up to encourage a collaborative approach between stakeholders, with representation from the professional specialist societies (cardiologists via the British Cardiovascular Intervention Society (BCIS), and cardiac surgeons via the Society of Cardiothoracic Surgeons (SCTS)), those involved in data collection and management (the former Central Cardiac Audit Database (CCAD) group, and representatives from the Department of Health (England), the National Specialist Commissioning Advisory Group, and the National Institute for Health and Clinical Excellence. A consensus document was produced outlining governance recommendations about the introduction of TAVI, including the requirement for a multidisciplinary team approach to treatment decisions, and a description of the infrastructure required for a centre wishing to run a TAVI program.1 Central to these suggestions was the requirement for all centres to collect procedural, outcome and follow-up data for every case, so that a National audit registry could be established, and that commissioning of procedures would be conditional on such data being collected.
Organisation and funding
The UK TAVI Group comprises four subgroups (online appendix 1): the Steering Group, the Data Management Group (DMG), the Clinical Research Group and the Dataset Group. The Steering Group provides overarching intellectual and professional leadership, and oversight of the developing UK TAVI programme. The DMG acts as custodians of the data, with responsibility for planning analyses and helping in the development of scientific manuscripts. The DMG also acts as a review panel for initial screening of academic requests for access to the TAVI dataset. The role of the Clinical Research Group is to develop and encourage academic analysis of the TAVI database, with plans to develop risk modelling specific to this new intervention. A grant from the Health Technology Assessment programme of the National Institute for Health Research was awarded in 2012 to fund a randomised UK trial assessing the clinical effectiveness and cost utility of TAVI. The Dataset Group devised the original TAVI dataset and is responsible for delivering new iterations and ensuring the change control process. A new dataset was created in June 2012, and will be collected from January 2013.
TAVI data collection was initially run centrally by the CCAD team, along with its other major UK cardiac audits. In 2011, CCAD became part of the newly established National Institute for Cardiovascular Outcomes Research (NICOR) (http://www.ucl.ac.uk/nicor/), which, in addition to the TAVI registry, also hosts a number of other national cardiovascular registries.2–5 NICOR, in turn, is part of the Centre for Cardiovascular Prevention and Outcomes in the Institute for Cardiovascular Science at University College, London.
Initial funding for the UK TAVI steering group was provided by the National Specialist Commissioners, for the purpose of data collection. As a part of the NICOR suite of cardiovascular audits, additional resource and funding now comes from the Department of Health. The costs of local data entry are borne by the participating hospitals.
Staff at NICOR provide telephone support via a help desk for technical issues and, together with the TAVI Steering Group members, respond to queries regarding case scenarios and definitions. A secure drop box can be used to analyse potential technical problems related to data uploads, file structures and field mapping errors.
The data: who is in the database, what is measured, how are individual patients identified?
Methods for data recording and transfer
The TAVI database is intended to record information about every TAVI procedure performed in the UK. A web browser-based interface has been developed to allow data entry and encrypted transfer to central servers at NICOR. This is available to all centres free of charge. For centres using their own database systems, all that is required is for these systems to generate a comma-separated-values file of a specified format. This can then be sent securely via the web browser interface to the NICOR servers.
For each TAVI procedure, a total of 95 variables are collected to the time of discharge from hospital. The current dataset is version 3.9 and is available at http://www.ucl.ac.uk/nicor/audits/tavi/dataset. The variables include patient demographic features, indications for TAVI, risk factors for intervention, details of the operators, technical aspects of the procedure, and adverse outcomes, including complications up to the time of hospital discharge. There are six additional fields provided for 1- and 3-year follow-up. Definitions of these variables are provided in the spreadsheet. In June 2012, the dataset group created a new dataset (version 4) which includes other features such as measures of frailty, and is compatible with the Valve Academic Research Consortium definitions.6 This has been distributed for collection starting in January 2013. It is available for download at http://www.bcis.org.uk/pages/page_box_contents.asp?pageid=694&navcatid=25. As of May 2012, there are approximately 3000 records in the UK TAVI database (figure 1).
Patient identification and tracking
The National Health Service (NHS) number provides a unique identifier for any person registered with the NHS in England and Wales. Mortality tracking is undertaken by the Medical Research Information Service using this unique patient identifier. It is a legal requirement for all deaths in the UK to be registered with Medical Research Information Service and it is not possible to effect any form of burial or cremation or similar process for the deceased without such registration. Thus, tracking yields very complete data on life status. The NHS number is pseudonymised within the database, and patient addresses are not recorded. In Scotland, a tracking system operates on date of birth and name only and is, therefore, less robust. No formal tracking exists in Northern Ireland. Other identifiers recorded in the database include hospital numbers, dates of birth and postcodes. Researchers do not have access to these aspects of the dataset, and hospital identity is also protected. Age at time of TAVI is provided along with number of days from procedure, to follow-up tracking (mortality or census date).
Data quality: accuracy, validation, completeness
The aim is for all TAVI procedures performed in any hospital in the countries of the UK (England, Scotland, Wales and Northern Ireland) to be recorded in the TAVI database. The UK TAVI Steering Group includes both Commissioners and healthcare professionals from across the UK, and so, all centres performing the procedure are known. The initial position statement from BCIS and SCTS in 20091 included the statement that ‘any unit performing the (TAVI) procedure has to provide procedural, outcome and follow-up data, (in the form of the agreed BCIS/SCTS dataset), to a centrally held Department of Health database for event tracking’. The commissioning framework in 20097 includes the following statement: ‘Mandatory collection of key data will be required from all UK centres in which the procedure is undertaken, in the form of a registry. The registry will include all new patients undergoing the procedure, as well as those who have already received it…continued funding of TAVI centres will be dependent on compliance with data collection and submission.’ Thus, strong professional and commissioning pressure was applied to encourage data collection. In addition, some of the initial funding from the commissioners for the TAVI group was ring-fenced to provide support staff at NICOR whose main remit was to liaise with all TAVI centres and their data entry personnel to assist with timely and accurate data entry. A data completeness report is sent regularly to all centres so that areas for improvement can be readily identified.
Initial publication efforts focused on the analysis of all data from the start of TAVI in the UK (2007) to the end of December 2009, which when once completed, was cleaned and locked.8 Subsequently, each calendar year's data have been cleaned and locked (but allowing updates to mortality tracking and other follow-up fields). To date, very high levels of completeness have been achieved, with only one hospital failing to participate fully. For data relating to procedures undertaken before the end of 2010, completeness of valid data was 99.6% for demographic data, 96.4% for risk factors, 97.4% for procedural variables and 98.5% for in-hospital outcomes. Mortality tracking was achieved in 100%.
Accuracy and validation
There is no external data validation, however, range checks are applied to appropriate fields. Missing and extreme values and data inconsistencies are queried by direct contact with the TAVI centre. Reliance is placed on local data entry and clinical staff to ensure data accuracy.
Use of the data
The main purpose of the TAVI data collection is to provide a detailed and accurate description of the way this evolving technology is being used to treat patients, to describe the results of this treatment and to be reassured that it is being undertaken as safely as possible in a cohort of patients in whom the natural history of their condition (severe aortic stenosis and significant comorbidity) raise them to high operative risk status. It is hoped that these data will ultimately improve the care of patients by guiding the therapy to those who will gain most benefit, and benchmarking TAVI units so that all can learn from the best practice of others. Nevertheless, the UK TAVI data collection process is relatively new compared with the other national UK registries. While there are plans to develop regular reports and feedback to institutions, this and many other developments are ongoing. In an era when commissioning of new technologies is likely to change, following establishment of the National Health Service Commissioning Board in 2013, it is hoped that comprehensive clinical and outcome data, such as that collected since the first TAVI procedure was performed, may be used to inform the safe introduction of other new technologies.
Both, large national registries and randomised controlled clinical trials will be needed to try to answer the many outstanding questions about this therapy. The TAVI Steering Group is responsible for deciding on a program of analysis of the growing UK TAVI dataset. In addition, a data-sharing agreement containing a data-governance framework has been created, and is available from the NICOR web site. Through this mechanism, the dataset is available to other research groups, under the guidance of the DMG. There are plans to expand the research potential of this dataset by cross-links with the other cardiac audits also hosted by NICOR.
Strengths and weaknesses of data
The main strengths are the inclusion of all consecutive patients treated in the UK, regardless of device manufacturer or access route, from the very start of the use of this new technology. Thus, patients of a range of risk profiles are included, and the numbers in the registry are large. The learning phases of all institutions have been captured, as has the changing patient case mix and evolving equipment. Tracking of all-cause mortality of the English and Welsh patients is extremely robust.
This data collection, however, shares the weaknesses of other national registry programs. There is always a balance between the size of the dataset and the willingness and ability of data entry teams to collect it accurately. In addition, making changes to the dataset risks losing collection from some units whose ability to modify data collection software is limited. Other than mortality tracking, the accuracy and completeness of the data are dependent on the individual centre's efforts, and other than range checks and checks for internal validity, there are no external validation processes in place. While we believe that centres make great efforts to submit accurate data, the lack of validation in such registries does constitute a weakness. Also, apart from life status, later clinical and quality-of-life follow-up is limited. Nevertheless, planned linkage with the other NICOR registries will allow determination of many important future events, such as recurrent need for later cardiac and cardiothoracic surgical interventions.
The TAVI registry is in its very early stages of evolution. While it has had an extremely encouraging start, there is clearly a long way to go before it becomes a mature tool, and there are many directions for future development. These include increased research, post-marketing surveillance, national and international reporting and comparisons of data, and the collection of data in the new and improved dataset version 4 that include frailty indices, to try to create an accurate risk-adjustment model. With cross-linkage of data, longitudinal tracking of a patient's treatment over time across a number of different subspecialties will be possible.
This approach of key stakeholders collaborating in the introduction of a new technology and therapeutic procedure (TAVI) within the NHS does, we believe, offer a template which could be modified and refined for wider use when changes in national commissioning in the UK occur in 2013. We would venture to suggest that it could contribute to a process of 'commissioning by evaluation', and help encourage the appropriate and safe introduction of new technologies more rapidly in the future.
The members of the UK TAVI Steering Group, Data Management Group, Dataset Group and Clincal Research Group.
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.
Files in this Data Supplement:
- Data supplement 1 - Online Appendix 1
Competing interests None declared.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.