Practical Cardiology Advisory Board member Dr. Brett Nowlan details his experiences with wearables as a practicing cardiologist.
A long-time patient of mine came into our office for a follow-up recently. He has been experiencing palpitations while hiking with his dogs, and we have been trying to make a formal diagnosis. Holter monitors have not been helpful, due to the intermittent nature of his symptoms. He arrived with data from his KardiaMobile 6L device, which pairs with his smartphone; and presented me with a stack of 6 lead EKG tracings he had obtained the prior weekend. This is now a common enough scenario in my practice to only require a brief acknowledgement.
Smartphones, and now smart watches, are big business, and widespread. In 2019 Pew research data showed that 81% of Americans own smartphones. The 65 years and older demographic has the lowest ownership of smartphones, but even still ownership reached 53%. Health tracking on these devices, and integration with wearable devices, are ubiquitous. Wearable devices alone are expected to have a market size of $70 billion by 2025. Arguably the most immediate impact that healthcare providers will notice with these devices, is in atrial fibrillation detection and management.
The Apple Watch has an ‘irregular rhythm’ algorithm, designed to detect atrial fibrillation. The Apple Heart Study correlated this algorithm with a wearable EKG patch. Of the over 400,000 monitored participants without a prior diagnosis of atrial fibrillation, 0.52% received notification of an irregular rhythm. Of those that then completed EKG patch confirmation, 34% were diagnosed with atrial fibrillation. Of those who were wearing both an iWatch and an EKG patch simultaneously, the iWatch had an 84% positive predictive value for observing true atrial fibrillation. This is impressive for a multifunctional device that lives on the wrist. The HEARTLINE Study is an ongoing randomized control trial assessing atrial fibrillation detection in patients aged 65 and older, using the Apple Watch.
This data is not limited to Apple products. The single-lead KardiaMobile device from Alivecor has users place their fingers on a recording pad, which communicates the tracing with a smartphone. This device has 87% sensitivity and 97% specificity for the diagnosis of silent atrial fibrillation, compared to a traditional 12-lead EKG.
Fitbit recently received FDA clearance for atrial fibrillation detection with their Fitbit Sense watch. Fitbit submitted multi-site clinical trial data for this clearance, involving 472 participants, demonstrating 98.7% sensitivity for detecting atrial fibrillation, and 100% sensitivity. The Fitbit Heart Study is an ongoing validation of its wearable technology in this regard.
Of course, this technology does not live in isolation. The data that is generated requires analysis, confirmation, and action from the healthcare team. This is yet another time demand in a busy office, and mostly one that cannot be financially compensated. Artificial Intelligence algorithms are rapidly developing to respond to this, and such networks are already in effect. Northshore University HealthSystem is currently teamed up with Carnegie Mellon University and physIQ for remote monitoring of CHF patients post-discharge, using patch-based wearable biosensors, and AI analysis and synthesis. The LINK-HF Study has shown this technology capable of predicting heart failure hospitalizations a median of 6 days in advance, with about 85% sensitivity and specificity.
All this collected data has a privacy concern attached to it. This is a moving target however, with younger generations clearly more at ease with this in concept and practice. Whether this remains true, as this data continues to evolve and increasingly leads to real-world consequences, remains to be seen.
Speaking of real-world consequences, we all have patients whose emotional health is negatively impacted far more than their physical health is helped, with this kind of information.
I have a 96-year old patient who I constantly have to steer away from measuring her own blood pressure, as each measurement gets worse as she becomes more anxious. She has attended walk-in clinics and emergency rooms on a number of occasions after these negative feedback loops. I would dread a wearable device on her wrist. The net impact of unregulated ‘screening’ is likely beneficial for most, but clearly psychologically harmful for some.
Back to my dog-walking patient—with documented symptoms and tracings in hand, he clearly has experienced his palpitations from isolated PVCs. Reassurance for him (and me), and a routine follow-up in six months. Whatever our personal thoughts or degree of buy-in to these devices; they are here to stay, are marketed directly towards patients, and will only become more ubiquitous.