Wearable technology can safely identify heart rate irregularities that subsequent clinical evaluations confirmed to be atrial fibrillation, according to a new study from the Stanford University School of Medicine and Apple now published in the New England Journal of Medicine (NEJM).
Atrial fibrillation, a type of irregular heart rhythm, is a leading cause of stroke and hospitalization in the United States, but due to its elusive and sporadic symptoms, the condition often goes undetected. With more than 400,000 participants enrolled in eight months, the Apple Heart Study is the largest virtual study to date.
“The study’s findings will help patients and clinicians understand how devices like the Apple Watch can play a role in identifying atrial fibrillation, a deadly and often undiagnosed disease,” said Mintu Turakhia, MD, associate professor of cardiovascular medicine, in a press release. “Additionally, these important findings lay the foundation for further research into the use of emerging wearable technologies in clinical practice and demonstrate the unique potential of large-scale app-based studies.”
Turakhia and Manisha Desai, PhD, professor of medicine and of biomedical data science, are the senior authors of the study. Marco Perez, MD, associate professor of cardiovascular medicine, is the lead author. The study chair is Kenneth Mahaffey, MD, professor of cardiovascular medicine.
The study was launched through a research sponsorship by Apple in November 2017 to determine whether software on the Apple Watch could use data from the Watch’s heart-rate pulse sensor to identify atrial fibrillation, which is one of the most commonly diagnosed significant cardiac arrhythmias in the United States, affecting up to six million people.
During the study, the researchers found that only 0.52% of participants received an irregular pulse notification, assuaging concerns about potential over-notification in healthy participants. Those who were flagged for an irregular pulse received follow-up care through a heart-monitoring technique called an electrocardiography (ECG) patch, which continuously monitors electrical impulses generated by the heart, for one week.
Of those who received a notification and were monitored by the ECG patch about two weeks later, 34% were found to have atrial fibrillation. Because atrial fibrillation is an intermittent condition, it’s not surprising for it to go undetected in subsequent ECG patch monitoring. Comparison between irregular pulse-detection on Apple Watch and simultaneous electrocardiography (ECG) patch recordings showed the pulse detection algorithm has an 84% positive predictive value.
During ECG patch monitoring, participants’ Apple Watches continued to monitor pulse irregularities. If a participant had an irregular pulse detected, 84% of the time this was confirmed to be atrial fibrillation on the simultaneous ECG patch. This, said Perez, demonstrates that the algorithm in the Apple Watch can successfully identify atrial fibrillation. Information from this study could be used to inform further clinical evaluation.
The study also showed how digital health alerts can enhance engagement with the health care system overall. A survey of participants who received an irregular pulse notification showed that 76% contacted either the telehealth provider or a non-study provider, suggesting that many actively sought medical attention as a result of an irregularity identified by their Apple Watch, said Turakhia.
“As the number of app-based health studies grows, developing additional methods to maximize self-report data accuracy and engagement will be an important area of investigation,” said Turakhia.
He added that the design of the study provided several valuable insights regarding patient engagement and interventions that will prove helpful in designing further studies. For one, the study’s completely virtual nature eliminated the need for patients to be physically present and allowed for the implementation of a massive recruitment strategy in a relatively short period of time. Also, because screening was done via interaction with the app, the study was administered at scale with little incremental cost.
“The performance and accuracy we observed in this study provide important information as we seek to understand the potential impact of wearable technology on the health system,” said Perez. “What the Apple Heart Study shows us is that atrial fibrillation is just the beginning. We can look ahead to other areas of preventive medicine. Further research will help people make more informed health decisions.”
To participate in the study, each individual needed an Apple Watch (series 1, 2 or 3) and an iPhone. (The Series 4 and 5 Apple Watches, which feature a built-in ECG, weren’t part of the study, as they were released after the study’s launch.)
The Apple Heart Study app intermittently checked the heart-rate pulse sensor for measurements of an irregular pulse. If it detected an irregular pulse, the participants received a notification and were asked to schedule a telehealth consultation with a doctor involved in the study through American Well. Participants were then sent ECG patches from BioTelemetry, which recorded the electrical rhythm of their hearts for up to a week.
Researchers from the Lankenau Heart Institute, Jefferson Medical College, the University of Colorado School of Medicine, Cooper Medical School of Rowan University, StopAfib.org, the American Foundation for Women’s Health and Duke University also contributed to the study.
Perez will present additional findings from the study in a session titled “Apple Watch App Identifies Clinically Important Arrhythmias Other Than Atrial Fibrillation: Results From the Apple Heart Study” at the American Heart Association 2019 Scientific Sessions on November 16, 2019 at 3:45pm.
