For investigators

All sites who enrolled patients into the original BAFTA study will be eligible for inclusion in this follow up. It is important that as many of the original participants as possible are followed up in this study. Therefore all eligible sites will be contacted by the research team.

Study data

Searches of the electronic medical records for the 1440 included patients (973 randomised and 467 non-randomised) will be carried out. Using MIQUEST data will be collected on: major vascular events; haemorrhages; use of warfarin (and time periods); and use of antiplatelet agents (and time periods) and other medications. Data on deaths will be obtained from the MRIS.

Patients who are no longer registered with their original practices will be traced via their NHS number. ‘Intention to treat’ and ‘on treatment’ analyses will be performed for randomised and non-randomised cohorts. These will compare all-cause mortality; stroke and vascular event rates; and haemorrhage rates.

Site workload

Participating sites will be required to run a MIQUEST search and send the results to the study team. It is expected that this will take a maximum of 3 hours of staff time, though in most cases this will be less. Full support is available from the research team.