The wearable health device market continues to grow. Though sports and fitness applications still dominate, more direct health care uses are predicted as researchers find out where they have the most value. 

For physicians already struggling to manage the tsunami of data streaming in from all sides, having an army of sensor-wearing patients pinging them with readings 24/7 could become a nightmare. A JAMA Viewpoint essay argues that medicine needs to take a lesson from aviation. 

Jet engines with 100,000 individual parts are monitored constantly with readings taken on temperature, pressures, shaft speeds and vibrations. Machine learning has been used for “novelty detection” when readings veer from established patterns and produce early warnings that unscheduled maintenance is needed.  

“The full potential of health monitoring for people will only be realized when individualized models underpin the monitoring algorithms,” the authors wrote. “Prospective validation that it promotes health, rather than exacerbates false alarms, will be vital.” 

Related Coverage

Privacy, security guidelines aim to bolster use of mHealth

The AMA is committed to making technology an asset rather than a burden, and developed a Digital Health Implementation Playbook that packaged key steps, best practices and resources to extend care beyond the exam room. 

The AMA also is a co-founder of Xcertia, a collaborative effort to establish and promote guidelines to improve the quality, safety and effectiveness of mobile health apps.  

JAMA Network™ journals have published peer-reviewed research that sheds light on uses for wearable digital health solutions. 

Smartwatches and stroke prevention. A JAMA Cardiology study found how to detect a leading cause of stroke—atrial fibrillation (AF)—by using a publicly available mobile app and a commercially available Apple Watch and then creating a machine-learning algorithm known as a “deep neural network.” 

University of California, San Francisco researchers used 139 million heart-rate measurements to “train” a deep neural network to compute the time between heartbeats. It outperformed two standard means of AF detection.  

Augmented Intelligence in Medicine

The AMA is committed to helping physicians harness AI in ways that improve patient care.

Patching in to quicker AF diagnosis. Patients wearing a self-applied electrocardiogram patch for two weeks at the start and two weeks at the end of a four-month period had a significantly higher rate of AF diagnosis than those not wearing a patch, according to a July 2018 JAMA study

Active monitoring was associated with earlier initiation of anticoagulant therapy and other AF-related interventions. 

Surgically precise outcome measures. The use of wristband pedometers and quality-of-life surveys provided objective and subjective functional-recovery measures for patients who had major abdominal cancer surgery. Step measures were taken before surgery, during hospitalization and after surgery to measure mobility.  

City of Hope researchers wrote in JAMA Surgery that many patients self-reported a return to pre-surgery quality-of-life levels even as their number of steps lagged. They suggested these results can be used to tailor postoperative care to detect complications and prevent readmissions. 

Interestingly, the first use of the term “wearable” in a JAMA Network™ journal appeared in the October 1933 edition of Archives of Ophthalmology, now JAMA Ophthalmology. It referenced “telescopic spectacles” that could be worn—as opposed to handheld “opera glasses.” 

Static Up
Featured Stories