AMLA-AF

  • Research type

    Research Study

  • Full title

    Application of machine learning algorithm to identify patients at highest risk of atrial fibrillation for targeted screening

  • IRAS ID

    293493

  • Contact name

    Pavidra Sivanandarajah

  • Contact email

    pavidra.sivanandarajah1@nhs.net

  • Sponsor organisation

    Chelsea and Westminster NHS Foundation Trust

  • Duration of Study in the UK

    1 years, 6 months, 5 days

  • Research summary

    Atrial fibrillation (AF) is the most common heart rhythm disturbance (arrhythmia). Individuals with atrial fibrillation have a five-fold increased risk of developing strokes. AF can be persistent or paroxysmal (intermittent). People with AF can be unaware that they have the condition as they may not display any symptoms. This results in many people presenting with AF at the time they present to hospital with a stroke. Strokes are debilitating, cause significant permanent disability and are a burden to the healthcare system.

    If AF is identified earlier, strokes can be prevented with treatment in the form of a blood thinner (known as anticoagulation). Currently, over 100 individuals will need to be screened in order to pick one individual with AF. However, a new machine learning algorithm has been developed (using previous AF risk score models and primary care data) and it has been predicted to increase our yield in identifying patients at risk of developing atrial fibrillation. Early data has suggested that it reduces the number needed to screen to 9 to pick one patient with atrial fibrillation.

    We will run this machine learning algorithm on the databases of general practices in the Hounslow area. Once the high-risk AF individuals have been identified, the general practices will be cluster randomised to four groups for further screening. The high-risk individuals will be recruited to compare the use of the latest mobile health technologies for AF screening as there are no comparative trials as yet. Due to the paroxysmal nature of AF, we will also use health technologies to determine the optimal duration and frequency of rhythm monitoring for AF screening as this is unknown.

    The data will be collated and analysed at West Middlesex University Hospital. We anticipate the study will last 18 months.

  • REC name

    London - Bloomsbury Research Ethics Committee

  • REC reference

    21/LO/0709

  • Date of REC Opinion

    2 Dec 2021

  • REC opinion

    Further Information Favourable Opinion