Atrial Fibrillation
Conditions
Brief summary
The purpose of the study is to confirm the software's ability to create a Lead-1 electrocardiogram (ECG) that is clinically equivalent to a reference device. Also, to confirm a rhythm classification algorithm and its ability to detect and classify heart rhythms into two categories (Sinus Rhythm or Atrial Fibrillation) using a single Lead ECG.
Interventions
OTHER1-Lead ECG
All participants will record three single-lead ECGs
DEVICE12-Lead ECG
All participants will simultaneously record three 12-lead ECGs
Sponsors
Iqvia Pty Ltd
Apple Inc.
Study design
Allocation
NON_RANDOMIZED
Intervention model
PARALLEL
Primary purpose
BASIC_SCIENCE
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
22 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Individuals who are 22 years of age and older * Able to read, understand, and provide written informed consent * Willing and able to participate in the study procedures as described in the consent * Have a wrist circumference that fits within the band * Able to communicate effectively with and follow instructions from the study staff * For subjects enrolled into the AF population, subjects must have a known diagnosis of AF and be in AF at the time of screening
Exclusion criteria
* Physical disability that precludes safe and adequate testing * Mental impairment resulting in limited ability to cooperate * Subjects with a pacemaker or implantable cardioverter-defibrillator (ICD) * Acute myocardial infarction (MI) within 90 days of screening or other cardiovascular disease that, in the opinion of the Investigator, increases the risk to the subject or renders data uninterpretable * Acute pulmonary embolism, pulmonary infarction, or deep vein thrombosis within 90 days of screening * Stroke or transient ischemic attack within 90 days of screening * Subjects taking rhythm control drugs * Symptomatic (or active) allergic skin reactions such as eczema, rosacea, impetigo, dermatomyositis or allergic contact dermatitis on both wrists or over electrode attachment sites * Known sensitivity to medical adhesives, isopropyl alcohol, watch bands, or electrocardiogram (ECG) electrodes including known allergy or sensitivity to fluoroelastomer bands primarily used in wrist worn fitness devices * A history of abnormal life-threatening rhythms as determined by the investigator * Significant tremor that prevents subject from being able to hold still * Pregnant women: Women who are pregnant at the time of study participation * For subjects enrolled into the sinus rhythm population, they must not have any diagnosis of AF
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of Participants With Software's Rhythm Classification of SR in Agreement to a Physician's Interpretation of a Gold Standard 12-lead ECG | 1 Day | Specificity of rhythm classification |
| Number of Participants With Software's Rhythm Classification of AF in Agreement to a Physician's Interpretation of a Gold Standard 12-lead ECG | 1 Day | Sensitivity of rhythm classification |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference | 1 Day | Number of ECGs that pass a visual overlay |
| Ease of Use | 1 Day | Average ease of use on a 1 (Unable to Use) to 5 (Easiest to Use) scale |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Atrial Fibrillation (AF) Patient with a known history of AF who is in AF at the time of study screening.
1-Lead ECG: All participants will record three single-lead ECGs
12-Lead ECG: All participants will simultaneously record three 12-lead ECGs | 301 |
| Normal Sinus Rhythm (SR) Patient with no known diagnosis of AF or other arrhythmia
1-Lead ECG: All participants will record three single-lead ECGs
12-Lead ECG: All participants will simultaneously record three 12-lead ECGs | 287 |
| Total | 588 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Protocol Violation | 0 | 14 |
Baseline characteristics
| Characteristic | Normal Sinus Rhythm (SR) | Total | Atrial Fibrillation (AF) |
|---|---|---|---|
| Age, Continuous | 59.5 years STANDARD_DEVIATION 17.73 | 66.9 years STANDARD_DEVIATION 15.47 | 73.8 years STANDARD_DEVIATION 8.62 |
| Race/Ethnicity, Customized Ethnicity : Hispanic or Latino | 57 Participants | 76 Participants | 19 Participants |
| Race/Ethnicity, Customized Ethnicity : Not Hispanic or Latino | 230 Participants | 512 Participants | 282 Participants |
| Race/Ethnicity, Customized Race: American Indian or Alaska Native | 0 Participants | 2 Participants | 2 Participants |
| Race/Ethnicity, Customized Race: Asian | 2 Participants | 3 Participants | 1 Participants |
| Race/Ethnicity, Customized Race: Black or African American | 46 Participants | 54 Participants | 8 Participants |
| Race/Ethnicity, Customized Race: Native Hawaiian or Other Pacific Islanders | 1 Participants | 1 Participants | 0 Participants |
| Race/Ethnicity, Customized Race: Not Reported | 3 Participants | 3 Participants | 0 Participants |
| Race/Ethnicity, Customized Race: White | 236 Participants | 528 Participants | 292 Participants |
| Region of Enrollment United States | 287 Participants | 588 Participants | 301 Participants |
| Sex: Female, Male Female | 159 Participants | 251 Participants | 92 Participants |
| Sex: Female, Male Male | 128 Participants | 337 Participants | 209 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 301 | 0 / 301 |
| other Total, other adverse events | 0 / 301 | 0 / 301 |
| serious Total, serious adverse events | 0 / 301 | 0 / 301 |
Outcome results
Primary
Number of Participants With Software's Rhythm Classification of AF in Agreement to a Physician's Interpretation of a Gold Standard 12-lead ECG
Sensitivity of rhythm classification
Time frame: 1 Day
Population: Classifiable Analysis Set: All subjects who had readable/classifiable paired Software and Reference strips were used. This analysis set was used for analyzing the primary endpoints of sensitivity and specificity.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Normal Sinus Rhythm (SR) | Number of Participants With Software's Rhythm Classification of AF in Agreement to a Physician's Interpretation of a Gold Standard 12-lead ECG | 236 Participants |
Comparison: H0: Sensitivity = 90% H1: Sensitivity \> 90% Under the assumption that the true population sensitivity is 95%, 231 subjects who were diagnosed with AF based on the 12-lead ECG reference strip and where the device algorithm classification produced a result of AF or SR was calculated to provide at least 80% power to reject the null hypothesis, using a one-sided type I error of 0.025. To obtain readable waveforms, a minimum of 260 subjects with a known diagnosis of AF were enrolled.p-value: <0.00011 sided exact binomial test
Primary
Number of Participants With Software's Rhythm Classification of SR in Agreement to a Physician's Interpretation of a Gold Standard 12-lead ECG
Specificity of rhythm classification
Time frame: 1 Day
Population: Classifiable Analysis Set: All subjects who had readable/classifiable paired Software and Reference strips were used. This analysis set was used for analyzing the primary endpoints of sensitivity and specificity.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Normal Sinus Rhythm (SR) | Number of Participants With Software's Rhythm Classification of SR in Agreement to a Physician's Interpretation of a Gold Standard 12-lead ECG | 238 Participants |
Comparison: H0: Specificity = 92% H1: Specificity \> 92 Under the assumption that the true population specificity is 96.5%, 226 subjects who were diagnosed with SR based on the 12-lead ECG reference strip and where the device algorithm classification produced a result of AF or SR was calculated to provide at least 80% power to reject its null hypothesis, using a one-sided type I error of 0.025. To obtain readable waveforms approximately 300 subjects with no known diagnosis of AF were enrolled.p-value: <0.00011 sided exact binomial test
Secondary
Ease of Use
Average ease of use on a 1 (Unable to Use) to 5 (Easiest to Use) scale
Time frame: 1 Day
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Normal Sinus Rhythm (SR) | Ease of Use | 1 = Unable to Use | 2 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 3 = Average Ease of Use | 14 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 2 = Below Average Ease of Use | 4 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 4 = Above Average Ease of Use | 41 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 5 = Easiest to Use | 240 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 4 = Above Average Ease of Use | 29 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 5 = Easiest to Use | 239 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 1 = Unable to Use | 1 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 2 = Below Average Ease of Use | 2 Participants |
| Normal Sinus Rhythm (SR) | Ease of Use | 3 = Average Ease of Use | 16 Participants |
Secondary
Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference
Number of ECGs that pass a visual overlay
Time frame: 1 Day
Population: Waveform Assessment Analysis Set: Paired Software and Reference strips collected from subjects were randomly selected. If 6 consecutive paired beats for analysis cannot be found in these strips, they were excluded from further analysis. This analysis set was used for assessing the quality of the clinical waveform.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Normal Sinus Rhythm (SR) | Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference | 60 Participants |
| Normal Sinus Rhythm (SR) | Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference | 65 Participants |
Comparison: H0: agreement proportion of visual display= 0.8 H1: agreement proportion of visual display\> 0.8 Under the assumption that the true population agreement proportion of visual display was 90%, 88 subjects would provide at least 80% power to reject the null hypothesis using a one-sided type I error of 0.05. To account for obtaining readable waveforms, approximately 140 subjects (70 SR; 70 AF) were randomly selected.p-value: <0.00011 sided exact binomial
Secondary
Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference
Difference in R-wave amplitudes between the software and gold standard reference
Time frame: 1 Day
Population: Waveform Assessment Analysis Set: Paired Software and Reference strips collected from subjects were randomly selected. If 6 consecutive paired beats for analysis cannot be found in these strips, they were excluded from further analysis. This analysis set was used for assessing the quality of the clinical waveform.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Normal Sinus Rhythm (SR) | Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference | 58 Participants |
| Normal Sinus Rhythm (SR) | Number of Participants With Software's Ability to Produce a Clinically Equivalent Waveform in Agreement to a Gold Standard Lead 1 Reference | 65 Participants |
Comparison: H0: agreement proportion of R wave amplitude = 0.8 H1: agreement proportion of R wave amplitude \> 0.8 Under the assumption that the true population agreement proportion of R wave amplitude was 90%, 88 subjects would provide at least 80% power to reject the null hypothesis using a one-sided type I error of 0.05. To account for obtaining readable waveforms, approximately 140 subjects (70 SR; 70 AF) were randomly selected.p-value: <0.00011 sided exact binomial