Skip to main content

Home

What is Atrial Fibrillation (AF)?

Atrial fibrillation (AF) is a heart condition that causes an irregular heartbeat. It is the most common disturbance of the heart rhythm and affects over a million people in the UK. Atrial fibrillation can cause palpitations, breathlessness and dizziness.

Often atrial fibrillation causes no symptoms and for many people atrial fibrillation comes and goes (paroxysmal AF). Because it can cause no symptoms and can come and go atrial fibrillation can be difficult to detect. It is estimated over 300 000 people in the UK are living with undiagnosed atrial fibrillation.

Atrial fibrillation increases the risk of stroke by approximately 5 times. About ten in 100 (10%) strokes happen in people unaware they have AF. There are medications available (anticoagulants) which can reduce the risk of AF-related stroke by 2/3rds if people are diagnosed early enough. 

  • Atrial fibrillation (AF) is the most common clinically significant abnormal heart rhythm in the population.
  • AF increases the risk of stroke, heart failure, cognitive decline and death.
  • It is currently estimated that over 1 million people are diagnosed with AF in the UK, and that over 300,000 people may have AF but not be diagnosed.

For further information about AF please see the NHS Choices website.

Why we need to detect AF earlier

  • Approximately 25-30% of people with AF are likely to be undiagnosed.
  • Around 20-30% of ischaemic strokes are caused by AF which can be severe, leading to disability and fatality.
  • Use of tablets (oral anticoagulant) can reduce the risk of a stroke due to AF by two-thirds.
  • Earlier detection of AF and treatment with oral anticoagulant can reduce the risk of stroke.

How FIND-AF works

FIND-AF is a computer tool that has been developed by the University of Leeds through funding from the British Heart Foundation. FIND-AF works in medical records held by GP practices. It uses age, sex, ethnicity and health problems recorded in these medical records to identify people more likely to have atrial fibrillation.

FIND-AF was developed and tested in over 2 million primary care records in the UK before this study.. When tested amongst people who have and have not been diagnosed with atrial fibrillation in the past, it was found to be highly predictive – in fact, amongst people defined as lower risk for atrial fibrillation by FIND-AF the chance of atrial fibrillation diagnosis in the next 6 months was only 1%.

In this study, FIND-AF will be used in medical records of selected GP practices across England. People who are identified as higher risk of atrial fibrillation are invited to use a handheld ECG recorder or wear a ECG recording patch for 2 weeks.

FIND-AF scores

FIND-AF is an algorithm that uses information held in primary care records to calculate the risk of a person developing atrial fibrillation (AF) in the next 6 months. It presents the average risk of people with the same risk factors as those entered for that person. It has been specifically developed by doctors and academics working in the UK National Health Service and is based on routinely collected data from thousands of GPs across the country who have freely contributed data to the Clinical Practice Research Datalink database for medical research. FIND-AF has been developed for the UK population, and is intended for use in the UK.
All medical decisions need to be in consultation with a healthcare professional. The authors and sponsors accept no responsibility for clinical use or misuse of this score. The science underpinning FIND-AF has been published in the Heart journal.

How FIND-AF is different to other algorithms

  • Algorithms can predict the risk of developing AF based on an ECG. However, ECGs are not usually available for most people in the community.
  • Algorithms can predict risk of developing AF using data available in primary care records – and were summarised in our systematic review published in the Heartjournal: https://heart.bmj.com/content/108/13/1020.abstract
  • There are a number of advantages of the use of FIND-AF over these algorithms:
    • Improved precision in a clinically-relevant prediction window:
      • FIND-AF was designed with input from clinicians and patients. Whereas other algorithms have prediction horizons that extend 5-10 years in the future, FIND-AF was designed to predict risk in the short term (6 months) so that it can inform clinicians of the merits of investigation and be meaningful to patients.
    • Widely applicable and inclusive:
      • FIND-AF can be applied to more primary care records than many previous algorithms because it was designed to not use types of data that are often not recorded in primary care records.
    • Designed to be used in clinical practice:
      • Whilst many algorithms have been developed for AF prediction, clinical implementation was the foremost principle during FIND-AF development. We ensured our design was implementable in primary care records.

Validated