A Study to Assess the Effectiveness of an Atrial Fibrillation (AF) Risk Prediction Algorithm and Diagnostic Test in Identifying Patients With AF. (PULsE AI)

A Randomised Controlled Trial for the Identification of Undiagnosed Atrial Fibrillation Patients Using a Machine Learning Risk Prediction Algorithm and Diagnostic Test

This is a trial to assess the effectiveness of an atrial fibrillation (AF) risk prediction algorithm and diagnostic test for the identification of patients with atrial fibrillation

Study Overview

• Status: Completed
• Conditions: Atrial Fibrillation

• Study Type: Observational
• Enrollment (Actual): 260

Contacts and Locations

Study Locations

• United Kingdom
  • Leamington Spa, United Kingdom, CV32 4RA
    • Local Institution
  • Ludlow, United Kingdom, SY8 2AB
    • Local Institution
  • Shropshire, United Kingdom, SY11 1RD
    • Local Institution
  • Warkwickshire, United Kingdom, B49 6QR
    • Local Institution
  • Wolverhampton, United Kingdom, WV10 8RN
    • Local Institution
  • Worcester, United Kingdom, WR1 2BS
    • Local Institution

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 30 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Probability Sample

Study Population

It is anticipated that approximately 18,000 patient records will be included in the trial.

It is anticipated that approximately 1,200 undiagnosed patients would be invited for AF diagnosis across all study sites.30 years is taken as the minimum age entry criteria as the algorithm was built on information from patients >30 years where AF becomes more prevalent.

Description

Inclusion Criteria:

Practice inclusion criteria for the trial are as follows;

• GP Practices within National Institute for Healthcare Research (NIHR) Clinical Research Network: West Midlands (CRN: WM) CRN: WM
• GP Practices using EMIS as their electronic medical record system of choice. Patient inclusion criteria for the trial are as follows;
• Patients registered at a participating practice, aged ≥30 years and without an AF diagnosis.
• Patients with a valid index date (see Section 3.3) Participant inclusion criteria for participation in the intervention arm are;
• As above, and who have provided written consent to: attend a research clinic, AF diagnosis check using a 12-lead ECG, and for access to medical records Participant inclusion criteria for participation in further AF remote diagnosis with an AliveCor Heart Monitor are;
• As above, and those with a negative or indeterminant ECG
• As above, and those with access to a smartphone

Exclusion criteria:

• Patients <30 years
• Patients with an existing diagnosis of AF
• Patients for whom the healthcare professional feels the study is unsuitable

Study Plan

How is the study designed?

Design Details

• Observational Models: Cohort
• Time Perspectives: Prospective

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort
Intervention arm; The AF risk prediction algorithm will be run on patient records within the Egton Medical Information Systems (EMIS) data base, in order to identify patients at risk of developing AF
Control arm; Patients may be diagnosed with AF through routine clinical practice only

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Percentage of participants with diagnosed Atrial Fibrillation during the research window in control and intervention arms	Prevalence of AF in the research window assessed	From the first collection of patient medical records at the beginning of the trial to the last collection of patient records following the intervention at the end of the trial (Up to 6 months)

Secondary Outcome Measures

Outcome Measure	Time Frame
Expected healthcare resource utilisation (Annual maintenance costs related to health states (informed by the primary endpoint), and daily treatment costs (including monitoring))	Up to 6 months
Quality-adjusted life years (QALYs)	Up to 6 months
Life years (LYs)	Up to 6 months

Collaborators and Investigators

• Sponsor: Bristol-Myers Squibb

Publications and helpful links

General Publications

• Hill NR, Arden C, Beresford-Hulme L, Camm AJ, Clifton D, Davies DW, Farooqui U, Gordon J, Groves L, Hurst M, Lawton S, Lister S, Mallen C, Martin AC, McEwan P, Pollock KG, Rogers J, Sandler B, Sugrue DM, Cohen AT. Identification of undiagnosed atrial fibrillation patients using a machine learning risk prediction algorithm and diagnostic testing (PULsE-AI): Study protocol for a randomised controlled trial. Contemp Clin Trials. 2020 Dec;99:106191. doi: 10.1016/j.cct.2020.106191. Epub 2020 Oct 19.

Helpful Links

• BMS Clinical Trial Information
• FDA Safety Alerts and Recalls

Study record dates

Study Major Dates

• Study Start (Actual): June 30, 2019
• Primary Completion (Actual): January 12, 2021
• Study Completion (Actual): January 12, 2021

Study Registration Dates

• First Submitted: July 23, 2019
• First Submitted That Met QC Criteria: August 2, 2019
• First Posted (Actual): August 5, 2019

Study Record Updates

• Last Update Posted (Actual): August 2, 2021
• Last Update Submitted That Met QC Criteria: July 29, 2021
• Last Verified: July 1, 2021

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Pathologic Processes; Heart Diseases; Cardiovascular Diseases; Arrhythmias, Cardiac; Atrial Fibrillation
• Other Study ID Numbers: CV185-703

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No