Assessment of Smartwatch to Identify Atrial Fibrillation

Study Questions:

Does a smartwatch application (app) identify atrial fibrillation (AF) during typical use?


The Apple Heart Study is a prospective, open-label pragmatic study sponsored by Apple. Participants included those aged >22 years who downloaded the app with an irregular pulse notification algorithm, downloadable from the Apple App Store between November 2017 and August 2018. All participants owned an iPhone and Apple Watch, resided in the United States, and were proficient in English. Those who reported prior AF or current use of anticoagulants were not eligible. All participants provided electronic consent. If a smartwatch-based irregular pulse notification algorithm identified possible AF, a telemedicine visit was initiated and an electrocardiography (ECG) patch was mailed to the participant, to be worn for up to 7 days. Surveys were administered 90 days after notification of the irregular pulse and at the end of the study. The main objectives were to estimate the proportion of notified participants with AF (30 seconds or more) shown on an ECG patch and the positive predictive value of irregular pulse intervals with a targeted confidence interval (CI) width of 0.10.


A total of 419,297 participants recruited over 8 months were included in the present study. Over a median of 117 days of monitoring, 2,161 participants (0.52%) received notifications of irregular pulse. Of these, 450 participants returned ECG patches with analyzable data. Average time of ECG application was 13 days after app notification of an irregular rhythm. ECG patches were worn for an average of 6 days. AF was present in 34% (97.5% CI, 29-39) overall and in 35% (97.5% CI, 27-43) of participants ≥65 years of age. Among participants who were notified of an irregular pulse, the positive predictive value was 0.84 (95% CI, 0.76-0.92) for observing AF on the ECG simultaneously with a subsequent irregular pulse notification and 0.71 (97.5% CI, 0.69-0.74) for observing AF on the ECG simultaneously with a subsequent irregular tachogram. Of 1,376 notified participants who returned a 90-day survey, 57% contacted health care providers outside the study. There were no reports of serious app-related adverse events.


The investigators concluded that the probability of receiving an irregular pulse notification was low. Among participants who received notification of an irregular pulse, 34% had AF on subsequent ECG patch readings and 84% of notifications were concordant with AF. This siteless (no on-site visits were required for the participants), pragmatic study design provides a foundation for large-scale pragmatic studies in which outcomes or adherence can be reliably assessed with user-owned devices.


As the authors note, the potential for pragmatic studies with wearable devices is great. Real-time monitoring without the cost of multiple study visits has the potential to reach more participants, resulting in granular data generalizable to many. However, we still have a ways to go, as this study points out. First, although these data support the positive predictive value of the app, the study cannot provide negative predictive value. Prehaps more importantly, over 1,000 people who did receive a notification of an irregular rhythm failed to initiate a study visit. Reasons for this are important to assess and will inform on how we use health-related apps in health care – for if we design a tool that patients do not use, we will not improve health care. Last, apps such as this one, together with wearables such as the Apple watch, have great potential. However, health and public health providers must make sure the benefits of such devices and programs are available to all irrespective of income, geographic residence, or race.

Keywords: Arrhythmias, Cardiac, Atrial Fibrillation, Electrocardiography, Pulse, Secondary Prevention, Telemedicine

< Back to Listings