In 1985, discussions took place between the UK Department of Health (then the Department of Health & Social Security) and representatives of the UK Society of Cardiothoracic Surgeons. Agreement was reached to set up a national registry for all artificial heart valves implanted through the National Health Service (NHS) hospitals within the UK. Three important decisions were reached :
The Registry office was set up in the Cardiac Surgical Unit of Hammersmith Hospital, London where the computerised database is still held. Each implanting NHS cardiac surgical unit was asked to nominate a 'link person' in that centre who would co-ordinate the centre's data collection, and be the contact person for the central registry office. A Freepost mail service was set up to facilitate transfer of registration forms.
From the outset the emphasis has been on achieving maximal co-operation from the UK centres and aim for completeness of patient registration. The data required for the Registration Form are therefore intentionally brief and simple to obtain. The form comprises patient identification details, surgeon and hospital identification and considerable detail on the implanted valve including each valves unique serial number. The Registry began data collection prospectively from 1st January 1986, with no attempt being made to obtain retrospective data. Patients are traced through the Office for National Statistics (England, Scotland & Wales) and the Central Services Agency (Northern Ireland) thus providing unique mortality information.
Although the Registry has historically collected data from NHS cardiac centres only, because of the overriding need to be able to identify all patients at risk after heart valve replacement surgery, the Registry now collects data from private cardiac surgical units throughout the UK.
The contribution of the UK Heart Valve Registry data to the knowledge of valve surgery is being increasingly recognised in scientific presentations and publications. Thus, the database is a unique information resource for patient demographics, short and long-term outcomes and heart valve performance.
Director: Professor Kenneth M Taylor
Ms Maria-Benedicta Edwards joined the staff of the UK Heart Valve Registry as its Manager in January 1996. She has a Batchelor's degree from the University of Liverpool and two Masters degrees from the University of London. Ms Edwards has undertaken a number of research posts in her career all in the field of health care services.
Ms Francesca Morgan joined the UK Heart Valve Registry at the end of February 1997 as Database Assistant. Prior to taking up this position she worked in primary health care services as a receptionist and administrative clerk. Francesca's responsibilities are data input and maintenance of the database as well as the role of liaison officer between the Registry and participating centres.
Mr John Clark joined the team of the UK Heart Valve Registry in January 1996. Unlike the other members of the Team, John contributes his services to the Heart Valve Registry on a part-time basis and is seconded to the Registry by the Department of Statistics and Epidemiology at the Imperial College School of Medicine, Hammersmith Campus. John's background is one of computer programming and statistics.
The newest member of the Registrys team is Ms Vanessa Thomas who joined in the summer of 1997. Vanessa, who is responsible for the secretarial and administrative functions within the Registry, has immense experience as a medical secretary in both the National Health Service and private health care sector.
The Patient Analysis Tracking System (PATS)
The UK Heart Valve Registry uses a database called the Patient Analysis & Tracking System (PATS) to record patient and valve details. The PATS system is a clinical information management tool which has allowed the Registry to create its own database based on its requirements. In addition to the storage of information, the PATS system is able to track and analyze patient data over time, conduct analysis of specific demographic or clinical data so that entire populations may be studied, tabulate data, compute averages and calculate patient survival curves. This flexibility and capacity to carry out such tasks is essential as the Registry is increasingly undertaking analysis on this population.
Collection of initial registry data
In the first ten years of the Heart Valve Registry's existence, the registration form and the method of inputting the data has remained unchanged. However, as a result of changing technology and the need to discover more about trends in heart valve disease and replacement surgery, the Registry introduced a modified version of the registration form in 1996.
Although the 'new' form contains more questions, its design is such that it will take less time to complete. A registration form is usually completed at the time of implant by a nominated member of staff at the implanting centre. However, this persons background is not necessarily clinical. Thus, the form has been designed with this in mind. Whoever completes the form is asked to shade the option(s) which corresponds to the answer or, to write only minimal information in the boxes provided. Each centre has received a set of guidelines on how to complete the form and are actively encouraged to contact the Registry for help should they have any problems or queries about how to do this. All participating centre are sent batches of blank forms upon request and each has its own unique hospital identifier number, known only to the Heart Valve Registry.
The completed forms are input into the database system manually on a daily basis. Regular and stringent validation and maintenance checks are carried out on the database to ensure correctness and completeness of data.
The Registry does not seek to obtain sequential follow-up data on patients apart from the occurrence of death and reoperation.
Re-operation
Re-operation is recorded by a new registration form completed for each valve operation subsequent to the initial registered valve replacement. Although the Registry data lack information on intercurrent events not leading to re-operation or death, this limitation was accepted from the planning stage as inevitable in order to maintain simplicity and completeness of data collection, rather than risk reduced co-operation and incomplete data from all UK implanting centres.
Notification of death
All patients entered into the database are 'flagged' for follow-up by either the Office for National Statistics (ONS : patients living in England, Scotland & Wales only) or the Central Services Agency (CSA : patients living in Northern Ireland). Among the many functions of these two agencies is to receive and process the death certificates for all persons dying within the UK.
At the end of every quarter (ie. last day of March, June, September and December) a list of patient details entered onto the Heart Valve Registrys PATS database system during the last quarter is sent to the ONS and CSA respectively. This is done in two different ways. The ONS' system is fully computerised and so, the Heart Valve Registry is able to download its last quarter's information onto disk and send this via the postal system. The CSA however, uses the more traditional card system which entails the Heart Valve Registry sending a batch of cards, each containing one patient's demographic details, to the CSA via the postal service.
Both agencies flag each valve replacement patient's details on the national mortality computer and thus, whenever a patient who has been flagged subsequently dies, death certification details are copied back to the Heart Valve Registry. The ONS notifies the Heart Valve Registry of a patient's death by sending to the office a copy of the certified death certificate. The CSA notify the Registry by returning the patient's card stamped with the patients date of death. This card is then sent to a second agency in Northern Ireland (Registrar for Deaths) and a request is made for a copy of the certified death certificate.
The link with these agencies therefore provides the Registry with the date and certified cause(s) of death, including date and place of death and post-mortem information (if carried out) for each heart valve replacement patient.
All participating centres can request at any time information about their own centre's performance. A centre may not however, receive any information about another centre's implant history or performance. This information can be provided either in paper format or on computer disk.
An annual report has been produced by the Heart Valve Registry since 1986 under UK Crown Copyright and has been distributed to the Department of Health as well as participating centres. The report provides information on the previous year's performance in terms of number of valves implanted, the sex and age of patients, the valve site and type ie. mechanical versus bioprosthetic. In addition, it offers some comparative analysis with previous years as well as reports on survival of patients within one calender year after implant.
A oral report has also been given each year at the annual meeting of the UK Society of Cardiothoracic Surgeons.
The UK Heart Valve Registry data is increasingly being used extensively by numerous authors of papers and, book chapters have cited UK Heart Valve Registry data with suitable acknowledgment.
Research output for the UK Heart Valve Registry :