Trial Outcomes & Findings for DEFINE AFib (Atrial Fibrillation) (NCT NCT04926857)

NCT ID: NCT04926857

Last Updated: 2025-08-01

Results Overview

Primary objective is to evaluate whether summary and episodic measurements collected by market-released LINQ ICMs are able to predict increased AF-related healthcare utilization (HCU), which are confirmed healthcare visits in the inpatient hospital, outpatient hospital, clinic/office, emergency department, or other care location (including remote visits) where AF was a reason or suspected reason for healthcare interaction. A predictive model was trained on 70% of the patients with adequate data available and assessed in a validation data set of the remaining 30% of patients, using rolling 60 day blocks. In each block, measurements in the first 30 days were used to predict whether confirmed HCU occurred in the trailing 30 days. Predictive value was assessed using sensitivity, specificity and other measurements within a confusion matrix as well as an area under the receiver operator curve, balancing the sensitivity and specificity. Outcome measure entered here will be AUROC.

• Recruitment status: TERMINATED
• Target enrollment: 973 participants
• Primary outcome timeframe: Through study completion (2 years of enrollment, up to 2.6 years of follow-up)
• Results posted on: 2025-08-01

Participant Flow

During study execution, there was a sponsor decision to end the study earlier, so the study enrolled far less than the 5000 enrollments specified by the protocol.

Participant milestones

Measure	Enrolled Patients Patients that completed a survey on the study App that confirmed that they met the inclusion/exclusion criteria of the study and that they read and agreed with the informed consent.
Overall Study STARTED	973
Overall Study COMPLETED	864
Overall Study NOT COMPLETED	109

Reasons for withdrawal

Measure	Enrolled Patients Patients that completed a survey on the study App that confirmed that they met the inclusion/exclusion criteria of the study and that they read and agreed with the informed consent.
Overall Study Withdrawal by Subject	40
Overall Study Death	9
Overall Study Study product no longer in use	44
Overall Study Inadequate data to address primary objective	16

Baseline Characteristics

DEFINE AFib (Atrial Fibrillation)

Baseline characteristics by cohort

Measure	Enrolled Patients n=970 Participants Patients that completed a survey on the study App that confirmed that they met the inclusion/exclusion criteria of the study and that they read and agreed with the informed consent.
Age, Categorical <=18 years	0 Participants n=5 Participants
Age, Categorical Between 18 and 65 years	276 Participants n=5 Participants
Age, Categorical >=65 years	694 Participants n=5 Participants
Sex: Female, Male Female	414 Participants n=5 Participants
Sex: Female, Male Male	556 Participants n=5 Participants
Race (NIH/OMB) American Indian or Alaska Native	5 Participants n=5 Participants
Race (NIH/OMB) Asian	7 Participants n=5 Participants
Race (NIH/OMB) Native Hawaiian or Other Pacific Islander	0 Participants n=5 Participants
Race (NIH/OMB) Black or African American	23 Participants n=5 Participants
Race (NIH/OMB) White	777 Participants n=5 Participants
Race (NIH/OMB) More than one race	22 Participants n=5 Participants
Race (NIH/OMB) Unknown or Not Reported	136 Participants n=5 Participants
Region of Enrollment United States	970 participants n=5 Participants
Reason for enrollment AF Management	584 Participants n=5 Participants
Reason for enrollment AF Ablation	181 Participants n=5 Participants
Reason for enrollment Cryptogenic Stroke	91 Participants n=5 Participants
Reason for enrollment Palpitations	20 Participants n=5 Participants
Reason for enrollment Suspected AF	94 Participants n=5 Participants

PRIMARY outcome

Timeframe: Through study completion (2 years of enrollment, up to 2.6 years of follow-up)

Primary objective is to evaluate whether summary and episodic measurements collected by market-released LINQ ICMs are able to predict increased AF-related healthcare utilization (HCU), which are confirmed healthcare visits in the inpatient hospital, outpatient hospital, clinic/office, emergency department, or other care location (including remote visits) where AF was a reason or suspected reason for healthcare interaction.

A predictive model was trained on 70% of the patients with adequate data available and assessed in a validation data set of the remaining 30% of patients, using rolling 60 day blocks. In each block, measurements in the first 30 days were used to predict whether confirmed HCU occurred in the trailing 30 days. Predictive value was assessed using sensitivity, specificity and other measurements within a confusion matrix as well as an area under the receiver operator curve, balancing the sensitivity and specificity. Outcome measure entered here will be AUROC.

Outcome measures

Measure

Training and Testing Partitions n=864 Participants The set of enrolled patients that had enough device follow-up for a lead-in portion and a prediction portion was used to develop a prediction model for increase healthcare utilizations. There were 864 total patients with sufficient device follow-up. Patients were partitioned into a training set (N=601) and a testing set (N=263). A predictive model was built with the training set and assessed for predictive value in the testing set.

Healthcare Utilization Where AF Was a Reason or a Suspected Reason for the Healthcare Interaction.

0.76 Proportion Interval 0.7 to 0.81

Adverse Events

Enrolled Patients

Serious events: 0 serious events

Other events: 0 other events

Deaths: 9 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Adverse event data not reported

Additional Information

Nirav Patel

Medtronic

Phone: 1 (800) 633-8766

Email: nirav.a.patel@medtronic.com

Results disclosure agreements

• Principal investigator is a sponsor employee
• Publication restrictions are in place