Skip navigation

Implementing EBM in general practice - antithrombotic AF treatment


Clinical bottom line

Audit can reveal patients who need treatment. Evidence can be used to explain the risks and benefits. Patient choice may limit the number who take up treatment.


A Howitt, D Armstrong. Implementing evidence based medicine in general practice: audit and qualitative study of antithrombotic treatment for atrial fibrillation. BMJ 1999 318: 1324-1327.


The study was a prospective audit carried out in a 13,000 patient practice in Kent. Computerised records were for patients with a diagnosis of AF, or who had been prescribed digoxin. Paper records were examined to confirm diagnosis. Where records were incomplete or unclear patients were invited to attend for review.

Patients were stratified by current risk of stroke. Those eligible were invited for a structured educational review to discuss antithrombotic treatment. Those taking warfarin for additional indications, or unable to give informed consent, or those to ill to participate were excluded. Those unable to come to the surgery were seen at home.

Patients were asked if they were aware that they were at increased risk of stroke, and what they though the risk to be. Information on stroke risk, and on the benefits and harms of warfarin, and of aspirin as an alternative, were given. They were screened for conditions associated with AF, offered echocardiography, and then attended a second interview to decide on antithrombotic treatment.


There were 132 patients with history of AF of whom 100 were eligible for warfarin (and 43 were taking it). Of these 16 were unable to consent, eight too ill to participate, and 16 with other indications for warfarin. Most of the remained (56/60) consented to be interviewed.

Of the interviewed group:

Most patients were unaware of the risk, and only two felt able to guess what their risk was. The common themes were:

  1. Many patients decided not to have warfarin because they did not see themselves at risk, while patients starting warfarin feared the effects of a stroke.
  2. Advanced age and impending death was an issue, but the response was varied and unpredictable.
  3. Attitudes to change in treatment were predominantly negative.


Implementing evidence-based medicine while taking into account patient preferences can lead to interesting conflicts. Individual appreciation of risk of a bad thing happening, and the consequences of that bad thing, vary greatly. This may limit the uptake of some effective preventative therapies.