Physician Tips: What Wearables Are Being Used Now For Healthcare

For Fibronostics

Third in a series about the development of wearable medical devices

One of the challenges of making wearable medical devices clinically useful is making sense of the deluge of data these devices can generate and more importantly extracting relevant and actionable information for clinical decision-making. (published site)

In a broad review of wearable medical devices published last September in Future Medicine, Jessilyn Dunn, Ryan Runge and Michael Snyder outline several of the clinical studies that have been undertaken to test the usefulness of medical wearable devices when paired with analytics platforms to analyze and evaluate data from a range of these devices. Below is a summary of some of the highlights from their review.

The analytic methods included machine learning to expand the use of wearables into clinical applications that range from acute health events, such as infection and inflammation, to chronic conditions such as cardiovascular disease, diabetes, and pulmonary disease.

Surprisingly, using such algorithms have shown that measurements from consumer watches could be used to predict heart and metabolic disease risk markers. In a study of 233 normal volunteers published in PLoS Biology, researchers integrated data from wearable sensors with lifestyle questionnaires, cardiac imaging, sphingolipid profiling, and clinical measurements of various heart and metabolic disease markers. The wearable data was found useful in identifying a class of sphingolipids (cell lipids) that are associated with obesity, diabetes, and heart disease.[i]

Wearables monitoring cardiovascular health

In the eHeart Study an analysis presented at the 32nd AAAI Conference, a collaboration between analytics software maker Cardiogram and the University of California San Francisco explored detecting clinical conditions data in 14,011 wearable-equipped participants. This team demonstrated early success in detecting Type II diabetes (85% accuracy), atrial fibrillation (97% accuracy), sleep apnea (90% accuracy), and hypertension (82% accuracy).[ii]

In a subgroup analysis published in May 2018 JAMA Cardiology, however, 51 patients enrolled in the eHeart Study who were undergoing cardioversion at UCSF and wore a smartwatch showed that while the smartwatch data did passively detect atrial fibrillation, it did so with much less specificity and sensitivity compared to 12-lead ECG. Even so, the researchers concluded that data from wearable devices coupled with sophisticated algorithms can be used to leverage AF management and rhythm assessment.[iii]

Wearable devices tested in hospital settings

Paring consumer wearables with hospital technology via wireless connectivity may be another way such devices may transform clinical and hospital care.

Recognizing the advantages of wireless connectivity of wearables, a Canadian feasibility study in the Journal of Intensive Care researchers outfitted 50 stable ICU patients with a Fitbit Charge® to monitor post-ICU convalescence over a 24 hour period. Heart rate, arrhythmia detection and sleep measurements from the Fitbit were compared against clinical gold standards. While specificity for the detection of tachycardia was high (98.8%), sensitivity was low to moderate (69.5%). The researchers concluded that wearable devices could generate useful data for the majority of patients while providing greater mobility and comfort.[iv]

Outside clinic monitoring for metabolic health

The first continuous glucose monitor (CGM) was FDA approved in 1999. Since then a number of advancements have been made with these small devices for people with diabetes. While early CGMs all required finger sticks for calibration the newer models no longer need fingerstick calibration and can be integrated with insulin pumps to adjust dosing, edging closer to an artificial pancreas.

Wearable cardiology monitoring closing in on clinical grade testing

Cardio patches, wrist devices and other cardiology wearable devices are edging closer to producing clinical grade ECG results. In an American Heart Journal study comparing ECG monitoring with the CAM® patch to Holter monitoring in 50 patients showed that the P-wave patch identified nearly double the number of management-altering rhythms compared to the Holter monitor.[v]

Among wrist devices ViCardio wristband threatens to replace the sphygmomanometer blood pressure cuff by directly monitoring beat-to-beat blood pressure, while the AliveCor’s KardiaMobile device makes an EKG as close as your smartphone or smartwatch.

Wearable sleep, neurology and mental health devices

The Oura ring sleep tracker compared with polysomnography has been shown to detect sleep states with 96% sensitivity and agreed with polysomnography 65% the time for light sleep, 51% for deep sleep and 61% for REM sleep.[vi]

A study involving 155 unmedicated youths monitored for three‐to‐five days with minute‐to‐minute belt‐worn actigraph data showed that objective measures of sleep, circadian rhythmicity, and hyperactivity were abnormal in bipolar disorder, suggesting that such wearable devices could be used differentiate subjects with bi-polar disorder from those with attention deficit hyperactivity disorder (ADHD).[vii]

Wearable pulmonary devices

In a recent validation study, 84 teens between the ages of 13 and 17 were monitored for 1 week during which the wearable ADAMM device collected audio data to assess the number of cough events, ultimately demonstrating an asthma detection accuracy of 71%.[viii]

While there is clearly a long way to go to achieve the accuracy and precision needed for clinical decision-making, such diverse examples of wearable devices coupled with sophisticated analytics shows the potential for these devices to produce sweeping change in the way we detect, diagnose and manage a range of diseases.

Fibronostics is committed to partnering with physicians and providers to improve patient care by offering the benefits of technology to improve lives,and deliver high-quality, life-improving disease education, evaluation and monitoring.

For more information contact us via email, or by phone at 1-888-552-1603.



Sources: Future Medicine, Wearable and the medical revolution

[i] Lim WK, Davila S, Teo JX, Yang C, Pua CJ, Blöcker C, et al. (2018) Beyond fitness tracking: The use of consumer-grade wearable data from normal volunteers in cardiovascular and lipidomics research. PLoS Biol 16(2): e2004285. https://doi.org/10.1371/journal.pbio.2004285

[ii] Ballinger, B.; Hsieh, J.; Singh, A.; Sohoni. DeepHeart: Semi-Supervised Sequence Learning for Cardiovascular Risk Prediction. AAAI Conference on Artificial Intelligence, North America, apr. 2018. Available at: <https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16967/15916>

[iii] Tison GH, Sanchez JM, Ballinger B, et al. Passive Detection of Atrial Fibrillation Using a Commercially Available Smartwatch. JAMA Cardiol. 2018;3(5):409–416. doi:10.1001/jamacardio.2018.0136

[iv] Ryan R. Kroll†, Erica D. McKenzie†, J. Gordon Boyd. Use of wearable devices for post-discharge monitoring of ICU patients: a feasibility study. Journal of Intensive Care20175:64 https://doi.org/10.1186/s40560-017-0261-9

[v] Warren M. Smith, Fiona Riddell, Morag Madon, Marye J. Gleva. Comparison of diagnostic value using a small, single channel, P-wave centric sternal ECG monitoring patch with a standard 3-lead Holter system over 24 hours. American Heart Journal, Volume 185, 2017, Pages 67-73, ISSN 0002-8703, https://doi.org/10.1016/j.ahj.2016.11.006.

[vi] Massimiliano de Zambotti, Leonardo Rosas, et al. The Sleep of the Ring: Comparison of the ŌURA Sleep Tracker Against Polysomnography, Behavioral Sleep Medicine,17:2, 124-136, DOI: 10.1080/15402002.2017.1300587

[vii] Faedda, G. L., Ohashi, K., Hernandez, M. et al. Actigraph measures discriminate pediatric bipolar disorder from attention‐deficit/hyperactivity disorder and typically developing controls. J Child Psychol Psychiatr, 57: 706-716. doi:10.1111/jcpp.12520

[viii] Rhee H, Belyea MJ, Sterling M, Bocko MF. Evaluating the Validity of an Automated Device for Asthma Monitoring for Adolescents: Correlational Design. J Med Internet Res 2015;17(10):e234 URL: https://www.jmir.org/2015/10/e234