Pandemic Outbreak Response with Wearables: Trending?

Photo by FitNish Media on Unsplash

Digital Epidemiology and Infectious Diseases

The goal of digital epidemiology is to understand the patterns and causes of disease and health dynamics in populations to gain insight into how to mitigate and/or prevent disease or promote health using digital data. The source of this digital data is usually generated outside the public health system and comes from three broad sources;

1. Medical data from electronic health records or insurance claims at the individual level
2. Participatory syndromic data in the form of self-reporting i.e. population health indicators which are available before confirmed diagnoses or laboratory confirmation
3. Non-health digital data generated from keyword searches, social media, mobile and wearable applications [1,2].

One area that digital epidemiology can benefit is the prevention of the spread of influenza and influenza-like viruses, specifically of pandemic nature because of the following characteristics:

“These pathogens transmit efficiently between humans, have sufficiently long asymptomatic infectious periods to facilitate the undetected movement of infected persons and have symptomatic profiles that present challenges for differential diagnosis (particularly in the early periods of infection).” [3]

Based on the challenges that these viruses present, the digital data source has to be able to provide real-time information to aid in the early detection of an infected person to be able to respond to outbreaks and prevent further transmission within a population [4].

Wearable Devices and Early Infection Detection

Wearable devices (wearables) have technological features that may be feasible for detecting infectious disease patterns among populations. They are a category of devices that are powered by microprocessors that can be worn as accessories, embellished on clothing or accessories or implanted in or tattooed on the user’s body and have the ability to send and receive data [5].

Previous research studies suggest that the use of wearable sensor data may help overcome the challenges of infectious disease spread as they are able to continue to track an individual’s physiological measurements, such as resting heart rate (RHR), physical activity, and sleep. The measurements have shown to be able to identify abnormal fluctuations indicative of an acute infection in real-time when compared to baseline measurements and prevent the lag-time experienced by traditional surveillance reporting by the U.S. Centre for Disease Control (CDC) [4].

Photo by Martin Sanchez on Unsplash

Narrowing the focus to our current reality, as vaccine trials are underway for the coronavirus disease (COVID-19) pandemic, it is critically important to enable technologies to prevent further spread of the disease through early detection of COVID-19 symptoms as it is highly efficient in person-to-person transmission [6].

Thus we can apply the concepts of digital epidemiology by using COVID-19 as a springboard for understanding the future of digital pandemic detection.

Epidemiological Modelling Using Digital Data

In the digital epidemiology space, social media data sources have previously been used to detect flu trends but studies suggest that the insights on flu detection gained from these sources were generally overestimated. The main challenge with social media sources is that there is a heightened activity from both an individual level and through general media [4]. More recently, Mishra et al. investigated whether wearable devices could be used for early detection of COVID-19 retrospectively [7]. The researchers took into consideration RHR (shown to track the spread of respiratory viruses), step count and sleep activity to detect deviations from the baseline that may indicate a COVID-19 infection. The variables used for this study were [7]:

• Self-reported survey of illness
• Diagnosis and symptom dates
• Illness severity and symptom type
• Wearable device data i.e. RHR, steps and sleep activity

The key findings of this study were that elevated RHR and heart rate over steps measurements, using unsupervised anomaly detection algorithms, were altered prior to symptom onset in 85% of individuals who tested positive for COVID-19 or other illnesses which is consistent with the presymptomatic illness phase. In addition, a real-time model based on cumulative statistics of heart rate showed that it is possible to detect 67% of infection cases at or before symptom onset [7].

This study shows that with the use of digital biomarkers obtained from wearable devices of individual physiological measurements and behavioural and self-reported data inputs, it is possible to detect early symptom onset of COVID-19. Thus, the advantage of digital epidemiology is that diverse new data sources can be aggregated and anonymized to aid in outbreak response by using epidemiological modelling techniques [8].

Wearables Market Trends and Industry Forecast

The global smart wearables market is estimated to triple from $216.18 million in 2019 to $960.97 million by 2027 [9].

The United States is leading the market, followed by Canada. Specific to disease and health tracking, the following consumer needs are driving the wearables market due to their ability to collect, track and communicate a variety of health data [9–12]:

1. Health consciousness and awareness among people driving the inclusion of medical features onto smart wearables such as heart rate fluctuation monitoring, ECG, pulse rate tracking, etc
2. A 47% increase of remote patient monitoring by hospitals that are providing wearables to patients with chronic diseases
3. COVID-19 forcing a shift to remote care and telehealth enabling wearable devices to track health-related statistics continuously and in real-time

Wearables Use Case for Digital Epidemiology

Digital epidemiologists, as the aforementioned definition suggests, aim to find alternative freely available data for disease monitoring. Health-related data sources such as health records and lab tests are a challenge for early-stage, real-time surveillance for emergency response and there is a need for detailed data to indicate disease clusters and spread. Since the coronavirus and influenza virus infections present asymptomatic infectious periods that only become apparent upon symptom onset such as fever, cough or shortness of breath, these early-stage interventions are required [11].

From a population-level perspective, one possible solution that has been suggested by researchers is for volunteers to share their sensor and clinical data as well as self-report their symptoms. By doing so, there is a possibility for early identification of asymptomatic, symptomatic and pre-symptomatic individuals [12]. Currently, studies are underway to help identify early detection possibilities via the study by the Stanford Healthcare Innovation Lab to explore whether data points collected from wearables could help researchers detect COVID-19 even before symptoms emerge [13] and by the Scripps Research Institute that will monitor your resting heart rate, sleep, and steps, and also allow you to record symptoms like fever or coughing [14]. Both of these studies are volunteer-driven.

Figure 1. Market share% in Q1 2020 of the wearable devices industry players. Adopted from Wearable Computing Devices Market 2020–2025 (Research and Markets)

Best-In-Class Wearable for Symptom Detection: Apple Watch

Figure 2. Smartwatches dominating the market for wearables in 2025 while limited growth for fitness trackers

Apple Inc. dominated the smartwatch market and lead in market share in the first quarter of 2020 (Figure 1.)[15] Additionally, shipments grew by 22% (Figure 2.) for Apple with North America and Europe showing the fastest growing uptake due to the growing demand of the health-conscious consumer and the release of the Apple Watch Series 5 model in 2019 [16, 17]. While, historically, the Apple Watch did not have the first-mover advantage as the first smartwatch on the market, it did take an early-mover advantage in getting FDA clearance for its electrocardiogram (ECG) app with the Series 4 Apple Watch. Apple has now also announced the addition of a blood oxygen tracker to the latest Series 6 version which could be useful for early COVID-19 symptom detection [19, 20]. Apple’s efforts thus far, show a continuous trend towards improving the capabilities of its smartwatch offerings by increasing the physiological information available to its users from a health perspective, providing further insight to the user into their overall wellness.

A study that is currently underway by Apple and the University of Washington Medicine researchers is to test their newly announced Apple blood oxygen measurement feature together with other physiological signals to identify people who are falling ill in an effort to stop further spread of infection [19]. Two other epidemiological efforts Apple is undertaking include firstly, the release of mobility data trends which may help inform governments and health authorities in new public policy decision making [21]. And secondly, to accelerate contact tracing, Apple and Google joined forces to enable the use of Bluetooth technology to help governments and health agencies reduce the spread of the virus [21]. Apple is also competing in the medical devices space since its ECG app received FDA clearance indicating a shift from solely wellness products.

The key issues that Apple and researchers are aware of in this space are primarily the trade-off between personal data privacy and public health as there is an urgent need and opportunity to use technology to monitor and slow the spread of COVID-19 [19]. Another limitation in order to more continuously monitor vital signs to inform early and real-time detection is that most devices have a limited battery-life and devices require charging which means that there will be missing data points during these periods [13].

Wearables Startup Spotlight

In the startup space currently, innovations are going beyond smartwatches to obtain the same or additional physiological signals that could help with early infection detection. One area that will witness significant growth is smart clothing as the sensor can be placed in various places from vests to undergarments [9].

Figure 3. Myant Inc’s Connected Mask (Concept Design) — Source: myant.ca

Myant, a Canadian startup company that is in this space, knits sensors into everyday textiles, giving them the ability to sense and react to the human body and collect as many data points as possible. One of their latest advancements is a connected mask (Figure 3.) that is embedded with biosensors and can collect data on emissions from the respiratory system via volatile organic compound (VOC) sensing, providing new ways to assess health and performance for the user [22, 23] with potential applications for COVID-19 symptom monitoring.

Their main value proposition over smartwatches is that their solutions integrate with daily living as is seen with their latest masks that are currently a mandatory part of our daily lives [23].

The company views their “Textile Computing™” solutions as a way of “…enabling the creation of a personal digital twin that is continuously updated via pervasive textile interfaces and optimized by artificial intelligence” [24]. Another unique aspect of Myant’s textile solution is that their biosensors can be embedded within any type of clothing making them adaptable and applicable to any need [22]. This startup demonstrates some of the emerging trends of health sensors as companies shift focus to more robust data collection via a non-disruptive means making a stronger case for personalized and insightful health tracking.

Figure 4. Oura Smart Ring — Source: ouraring.com

Another wearable that has been used in studies to detect early infection of COVID-19 is a health tracker in the form of a ring i.e. the Oura Ring (Figure 4.)that tracks many of the same vitals as existing fitness trackers [25]. The TemPredict study by the University of California, San Francisco (UCSF) aims to build an algorithm to identify COVID-19 disease patterns with the use of digital biomarkers collected by the Oura Ring and symptom surveys for both frontline workers and the general population [26]. The Oura Ring also claims to be non-disrupting to daily life in comparison to smartwatches and it is also able to establish a more individualized baseline to indicate deviations. While “The Oura Ring is the personal health tracking device to beat in 2020” but it is an expensive wearable with prices as competitive as an Apple Watch [25].

The emerging technology companies that have existed pre-COVID are now getting involved in studies and developing products or pivoting their product benefits towards the detection of early signs of infection in the same way that established wearable companies are. The startup companies in this space will have to also show further validation through rigorous research studies as regulated products if they are to stay competitive within the market e.g. being classified as a medical device to appeal to consumers interested in advanced health tracking [18].

Wearables and Mobile Apps FDA Regulatory Pathway

Recently, the FDA has issued guidance that enables the innovation of the wearable device space including fitness trackers and mobile health applications [11]. A distinction to note is that while there will be software in the form of a mobile medical app, accompanying hardware i.e. the wearable itself is considered a stand-alone technology and may/may not function as a medical device. In 2018, Apple received clearance from the FDA for an upgrade to its watch by including a personal electrocardiogram, this app can classify whether there are signs of atrial fibrillation (AFib) or not. The app was considered moderate-risk and is classified as a Class II device via the 510K pathway [27]. With the added features that existing wearables are building on and startups are developing, generally, the regulatory pathway that is taken requires rigorous clinical studies and human factor testing to show that the user can use the medical drive correctly, read and interpret the results [28, 29].

With respect to COVID-19, there are two potential pathways a wearable technology can take depending on its intended use. If a product related to early detection meets the definition of a medical device, it can take one of two pathways (shown in Figure 5.).

Figure 5. Possible pathways for wearables and associated mobile medical applications.

1. Usual premarket pathway based on the risk class (top branch)
2. Emergency Use Authorizations (EUAs) are issued with the belief that the device may be effective in treating, diagnosing or preventing COVID-19 [30] (bottom branch)

From a disease surveillance perspective, the FDA generally does not regulate products that are intended to track locations or contacts associated with public health surveillance because these products generally do not meet the definition of a medical device [11].

Trend Forecast: Wearables for Digital Epidemiology in the next 5 years

At the beginning of 2020, slow growth was seen in the wearables space as companies suffered due to supply chain and store closures disruptions. But, it is still predicted that the market will grow over the next 5 years and shipments will grow two-fold by 2024 for wearable devices to about 203 million smartwatch units and 100 million fitness tracker units [31]. Both these areas will support the proliferation of digital data and encourage epidemiological insights related to infection.

To better understand user adoption of wearables for infectious disease detection, a 10-question survey was conducted using SurveyMonkey and distributed anonymously via social media and social networking platforms [32].

Key survey results are summarized below:

-Respondent age range 18–54 years old with 55% 25–34 years old
-90% of respondents live in North America
-47.5% of respondents own a smartwatch, 20% own a fitness tracker while the -remainder have none
-70% of respondents are more likely to purchase/use a wearable if it could inform them about early signs of an infection
-67.5% are more likely to share their data if it could contribute to early outbreak response
-Top 3 wearable concerns with wearables are: Price, Data Privacy and Too Many Devices
-Top 3 most important features: Functionality, Quality and Price

These results confirm that there is an interest from a user perspective to provide digital data from wearable and contribute to the public health goal of identifying early infection among the populous that could help mitigate disease spread. Not surprisingly, data privacy has been identified as a barrier to adoption by companies but through this survey, there is an indication of willingness to share data for outbreak response efforts.

Barriers to Adoption

Finally, while evidence from research studies described suggest there is an opportunity for early infection detection using the existing wearable technology infrastructure, the following barriers will need to be overcome in order for digital epidemiology to gain a stronghold in outbreak response and thus, successful technology adoption of wearables over the next 5 years:

• Data privacy as user data is generally private and requires data protection. Data will need to be aggregated, anonymized and decentralized [8]
• Data validation of input and output data through the use of both data-driven models and domain-specific knowledge instead of one approach only which will lead to better predictive capabilities [8]
• Non-disruptive user design for passive monitoring of a person’s vital signs that can be integrated into daily living and provide feedback as well as comfort, ease of use and aesthetic appeal [22]
• The regulatory pathway for continuous learning algorithms, currently not cleared by the FDA [28]
• Accessibility from an ethical and price perspective of wearables to improve health equity by promoting access to connectivity to enhance pandemic response [8]
• Algorithms that are able to differentiate symptoms of different respiratory illnesses [7]

The drivers and blockers for this trend analysis for the use of wearables for digital epidemiology and pandemic outbreak response have been summarized below:

Trend Drivers

Rapidly scalable

Ability to be less reactive and more proactive

FDA guidance and recent decision history

Device agnostic digital epidemiology

Data privacy and consumer willingness to share data

Health-conscious consumer-driven demand

Growth of the wearable device market

New Innovation away from Blue Chip

User acceptance

Trend Blockers

Data privacy

Accessibility

Price

Reimbursement pathway

Research studies

References

1. Salathé, M. Digital epidemiology: what is it, and where is it going? Life Sciences, Society and Policy 14, (2018).
2. He, Z. et al. A New Era of Epidemiology: Digital Epidemiology for Investigating the COVID-19 Outbreak in China. Journal of Medical Internet Research 22, (2020).
3. Madhav N, Oppenheim B, Gallivan M, et al. Pandemics: Risks, Impacts, and Mitigation. Disease Control Priorities: Improving Health and Reducing Poverty.The International Bank for Reconstruction and Development / The World Bank 3, (2017).
4. Radin, J. M., Wineinger, N. E., Topol, E. J. & Steinhubl, S. R. Harnessing wearable device data to improve state-level real-time surveillance of influenza-like illness in the USA: a population-based study. The Lancet Digital Health 2, (2020).
5. Hayes, A. The Ins and Outs of Wearable Technology. (2020). Available at: https://www.investopedia.com/terms/w/wearable-technology.asp (Accessed: 30th October 2020)
6. Islam, M. M. et al. Wearable Technology to Assist the Patients Infected with Novel Coronavirus (COVID-19). SN Computer Science 1, (2020).
7. Mishra, T. et al. Early Detection Of COVID-19 Using A Smartwatch. medRxiv (2020). Available at: https://www.medrxiv.org/content/10.1101/2020.07.06.20147512v1 (Accessed: 1st November 2020)
8. Tarkoma, S., Alghnam, S. & Howell, M. D. Fighting pandemics with digital epidemiology. EClinicalMedicine 26, 100512 (2020).
9. Smart Wearables — Global Market Outlook (2019–2027). Research and Markets — Market Research Reports — Welcome Available at https://www.researchandmarkets.com/reports/5116366/smart-wearables-global-market-outlook-2019 (Accessed: 1st November 2020)
10. Wearable Computing Devices Market — Growth, Trends, Forecasts (2020–2025). Research and Markets — Market Research Reports — Welcome Available at: https://www.researchandmarkets.com/r/6ykysp (Accessed: 1st November 2020)
11. Center for Devices and Radiological Health. Digital Health Policies and Public Health Solutions for COVID-19. U.S. Food and Drug Administration Available at: https://www.fda.gov/medical-devices/coronavirus-covid-19-and-medical-devices/digital-health-policies-and-public-health-solutions-covid-19 (Accessed: 1st November 2020)
12. Covid-19 pandemic may give wearables necessary push to become healthcare mainstays. Medical Device Network (2020). Available at: https://www.medicaldevice-network.com/comment/covid-19-pandemic-wearables-push-healthcare-mainstays/ (Accessed: 1st November 2020)
13. Quer, G. et al. Wearable sensor data and self-reported symptoms for COVID-19 detection. Nature Medicine (2020). doi:10.1038/s41591–020–1123-x
14. Stanford COVID-19 Wearables Project. Stanford Healthcare Innovation Lab Available at: https://innovations.stanford.edu/wearables (Accessed: 2nd November 2020)
15. This might be an early warning system for viral illness. detect Available at: https://detectstudy.org/ (Accessed: 2nd November 2020)
16. Lim B. et al. Global Smartwatch Market Revenue up 20% in H1 2020. (2020). Available at: https://www.counterpointresearch.com/global-smartwatch-market-revenue-h1-2020/ (Accessed: 3rd November 2020)
17. IDC: Hearables market sees growth, but smartwatch sales decline in Q1 2020. Gizmochina (2020). Available at: https://www.gizmochina.com/2020/05/29/idc-hearables-market-sees-growth-but-smartwatch-sales-decline-in-q1-2020/ (Accessed: 3rd November 2020)
18. 5 Hot Trends in Wearable Medical Devices. mddionline.com (2020). Available at: https://www.mddionline.com/digital-health/5-hot-trends-wearable-medical-devices/gallery?curr=29 (Accessed: 2nd November 2020)
19. Bishop, T. Can a watch detect COVID-19? Here’s how Apple and the University of Washington plan to find out. GeekWire (2020). Available at: https://www.geekwire.com/2020/can-watch-detect-covid-19-heres-apple-university-washington-plan-find/ (Accessed: 2nd November 2020)
20. How to use the Blood Oxygen app on Apple Watch Series 6 Available at: https://support.apple.com/en-ca/HT211027 (Accessed: 2nd November 2020)
21. Apple makes mobility data available to aid COVID-19 efforts. Apple Newsroom (2020). Available at: https://www.apple.com/ca/newsroom/2020/04/apple-makes-mobility-data-available-to-aid-covid-19-efforts/ (Accessed: 2nd November 2020)
22. Vision. Myant (2020). Available at: https://myant.ca/vision (Accessed: 2nd November 2020)
23. Hannah, P. Myant Unveils Connected PPE Concepts, Creating New Ways to Assess Health and Performance as Part of the Skiin Interconnected System of Biometric Garments. Myant (2020). Available at: https://myant.ca/myant-unveils-connected-ppe-concepts-creating-new-ways-to-assesshealth-and-performance-as-part-of-the-skiin-interconnected-system-of-biometric-garments/ (Accessed: 2nd November 2020)
24. Hannah, P. Myant to Unveil a World Where Human Connectedness is Transformed by Textiles at CES 2020. Myant (2020). Available at: https://myant.ca/myant-to-unveil-a-world-where-human-connectedness-is-transformeD-by-textiles-at-ces-2020 (Accessed: 2nd November 2020)
25. Etherington, D. Oura Ring health and sleep tracker review. TechCrunch (2020). Available at: https://techcrunch.com/2020/08/12/the-oura-ring-is-the-personal-health-tracking-device-to-beat-in-2020/ (Accessed: 4th November 2020)
26. UCSF TemPredict Study Available at: https://techcrunch.com/2020/08/12/the-oura-ring-is-the-personal-health-tracking-device-to-beat-in-2020/ (Accessed: 3rd November 2020)
27. Electrocardiograph Software For Over-The-Counter Use Product Classification. accessdata.fda.gov Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPCD/classification.cfm?ID=1007 (Accessed: 3rd November 2020)
28. Gerke, S., Shachar, C., Chai, P. R. & Cohen, I. G. Regulatory, safety, and privacy concerns of home monitoring technologies during COVID-19. Nature Medicine 26, 1176–1182 (2020)
29. Center for Devices and Radiological Health. Software as a Medical Device (SAMD): Clinical Evaluation — Guidance. U.S. Food and Drug Administration Available at: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/software-medical-device-samd-clinical-evaluation (Accessed: 3rd November 2020)
30. Emergency Use Authorization. U.S. Food and Drug Administration Available at: https://www.fda.gov/emergency-preparedness-and-response/mcm-legal-regulatory-and-policy-framework/emergency-use-authorization (Accessed: 2nd November 2020)
31. Koytcheva, M., Portela, R., McQuire, N. & CCS Insight. Wearables Market Proves Resilient to Threat of Covid-19. CCS Insight (2020). Available at: https://www.ccsinsight.com/press/company-news/wearables-market-proves-resilient-to-threat-of-covid-19/ (Accessed: 3rd November 2020)
32. Survey Monkey survey results conducted by the author