Wearable Devices and Applications for Heart Disease and Health

Jia Shen MD, MPHAssistant Professor of MedicineUC San Diego Health System

Disclosures

I have no conflict of interest related to products presented in this talk.

Overview of Talk

§ Introduction

§ Review of Wearable Technology and Applications:§ Activity Monitors/Health Applications§ Sleep Apnea Monitoring§ Arrhythmia Monitoring

§ Controversies/Summary

• Digital Health Market is expected to double by 2022 to >27 billion USD (Forbes).

• Digital Health Action Plan (2017): FDA’s vision for fostering innovation while protecting and promoting patient health and safety.

• September, 2017: FDA’s announces pre-certification pilot program highlighting software companies working in the digital health field – Apple, Fitbit, Johnson & Johnson.

Introduction

JAMA Cardiol. 2016;1(2):239. doi:10.1001/jamacardio.2016.0354

Fitness Trackers/Activity Monitors

§Additional sensors monitor heart rate, blood pressure, blood oxygen levels, temperature, perspiration. §Information can be integrated with apps and shared with physicians, friends, and family.

§ Measures: footsteps taken, distance travelled, type of movement, quality, and duration of sleep.

Fitness Trackers/Activity Monitors

Accuracy of Activity Monitors

JAMACardiol. 2017;2(1):104-106.doi:10.1001/jamacardio.2016.3340

§ Wrist worn heart rate monitors were inferior to chest electrode monitors.

§ Variability increased with moderate and vigorous exercise.

Accuracy of Activity Monitors

§ Variability in step count/distance accuracy and repeatability between devices.

§ Accuracy is affected by level and type of activity.

Kaewkannate,K.&Kim,S.BMCPublicHealth 16, 433(2016).https://doi.org/10.1186/s12889-016-3059-0

Data on Fitness Trackers

471 patients randomized to standard (telephone calls, text prompts, access to website resources) or enhanced (wearable device + web interface to monitor diet and activity) groups.

Increased weight loss with standard therapy at 24 months {-2.4 kg 95% CI: 1.0-3.7}

JAMA. 2016;316(11):1161-1171. doi:10.1001/jama.2016.12858

• Wearableactivitymonitorsarepopularandeasilyintegratedintoavarietyofhealthapplications.

• Theycanfacilitategoalsetting,monitoring,anddatasharing.• Possiblyincreasemotivationviasenseofcommunity,

rewards/goalsetting,competition.• Inselectmotivated patientsthiscanencouragedietaryand

lifestylechanges,physicalactivity,weightloss.• Manyprivateinsurancecompaniessubsidizethecostofan

activitymonitor/healthapplication.• Physiciansandhealthcaresystemsarefiguringouttherole

thesedevicescanandshouldplayinpatientcare.

Conclusions

Sleep Apnea Monitoring§ Prevalence of Sleep Apnea in the USA ranges between 9% to

38%. 4 in 5 people with sleep apnea remain undiagnosed.

§ Polysomnography is the gold standard: § Brain waves

§ Eye movements

§ Heart rate§ Breathing pattern

§ Blood oxygen level

§ Body position

§ Snoring

§ Limb movement§ Chest and abdominal movement

Wearable Sleep Apnea Monitors

UCSF Study with 6,115 users of the Cardiogram app for Apple Watch.

All wearables (Apple Watch, Fitbit, Garmin, Android) can detect Sleep Apnea when paired with an AI neural network- called DeepHeart.

90% sensitive in identifying sleep apnea in participants but only 60% specific.

Tison,GH.,etal.Circulation 136.suppl_1 (2017):A21042-A21042

Evolution of Sleep Apnea Monitoring

Arrhythmia Monitoring

§ Spectrum of options ranging from ECG marketed to consumers (Alivecor/Apple) to medical grade monitors (Preventice, Ziopatch, Biotel).

AliveCor/Kardia

Apps also can integrate with omron blood pressure cuff

Study of 1001 patients with mean CHADS2-VASc score of 3

Halcox,JPetal.Circulation 136.19(2017):1784-1794.

Apple Watch Series 4-6

Apple Watch Series 4-6

§ 419,297participantsfollowedmedianof117days.§ Series1-3usesopticalsensorstodetectirregularpulses+

proprietaryalgorithmtodetectAF.§ 7-daypatchECGmonitormailedtoptswithpossibleAF.§ 0.52%notified,AFwaspresentin34%.§ Positivepredictivevalueof0.84[95%CI,0.76-0.92].

Apple Watch Series 1-3

N Engl J Med. 2019 Nov 14;381(20):1909-1917. doi: 10.1056/NEJMoa1901183. PMID: 31722151.

§ 50ptspost-opfromcardiacsurgerymonitoredontelemetry.§ Series4+useselectrodestogenerateasingle-leadECG

savedtoPDFforinterpretation.

Apple Watch Series 4+

Seshadri, DR., et al. Circulation 141.8 (2020): 702-703.

Event Monitors/Mobile Cardiac Telemetry Devices

§ DeepNeuralNetworks(DNNs)aremodelsofstackedtransformationsthatlearntasksbyexamples.

§ DNNtrainedondatasetof>2millionECGs.§ DNNoutperformedcardiologyfellows,residents,andmedical

studentsindetectingfirstdegreeAVB,RBBB,LBBB,AF,sinustachycardia.

Nature Communications volume 11, Article number: 1760 (2020)

Controversies§ Hundreds of wearable devices are now commercially

available and marketed directly to patients/consumers. § Accuracy – high quality data? § Healthcare Cost – who pays?

§ Responsibility – who is responsible for healthcare data? § Integration – how do we integrate data in medical records? § Can these devices improve individualized patient centered

care and lead to reduced cardiovascular outcomes?§ Lack of short and long-term outcome trials.

Summary§ Virtual explosion of consumer based wearable medical

devices and applications.§ Benefits include: better design, increased availability, data

collection, and patient participation. § Use of these tools need to be driven by good clinical trials

and outcome data, which are currently lacking.§ Physicians, insurers, and health systems need to determine

the role of wearable technology in guiding patient care. § Evolution of medical devices and patient monitoring.