Wearable Devices and Their Effect on Sleep Medicine
Pradeep Sahota, MD
University of Missouri School of Medicine
University of Missouri Health Care
Sahota P. Wearable devices and their effect on sleep medicine [published online March 27, 2020]. Neurology Consultant.
We spend one-third of our life sleeping. And lately sleep hygiene has gotten more recognition among the general population given its impact on health and well-being. There is also a desire to understand patients’ individual quantity and quality of sleep. Simultaneously, the development of affordable electronic sensing devices that can be worn directly on the body (wearables) or placed near the body (nearables) to measure physiologic variables has also played a role in highlighting the importance of good health and sleep hygiene. Hence, wearable and nearable devices that track surrogate markers that may provide an approximation of the quality and quantity of sleep are becoming more popular among the general population.
Wearables and Nearables: An Overview
Wearables can measure one or more physiological variables such as activity and heart rate. In contrast, nearables can monitor several physiological and other variables (such as movement of the bed), and the information is exported via a Bluetooth or Wi-Fi signal to a cell phone, tablet, or computer.
For example, actigraphy uses an accelerometer to sense motion and provides information regarding quantity of sleep. Actigraphy has been researched and validated to provide a very close estimation of sleep quantity, although it is not fully accurate. Much less information is available regarding sleep stages or the quality of sleep. However, most wearables and nearables are not researched or fully validated. In addition, some of the software algorithms that determine whether the person is asleep are proprietary, and hence the validity of the measures projected by the device are based on incompletely understood determinants. Some devices may monitor heart rate in addition to the movement and may use clues from the regularity, as well as the rate, to suggest depth of sleep.
Measuring Sleep Quality
Devices that health care providers use to diagnose sleep apnea are approved by the US Food and Drug Administration (FDA). However, the quality of wearables used by the general population varies, even as newer devices with updated software enter the market. Most of them are not FDA-approved.
Algorithms are proprietary as to how these devices measure the quantity and quality of sleep. Information available regarding quantity of sleep, while not very accurate (high sensitivity but low specificity), provides a relative picture of night-to-night sleep quantity and variability. As such, it brings patients’ focus to the need for quality sleep on a regular basis. Specifically, patients with chronic insomnia may feel relieved if it is found that they do have a particular amount of sleep on a nightly basis.
Sleep diaries, which have been used for subjective understanding of the patient’s sleep, may not be as accurately filled out or diligently maintained on a day-to-day/night-to-night basis. In such circumstances, these devices may provide some additional data that can be useful. Actigraphy has been used in this context. A study conducted at the University of Michigan required patients to wear Apple Watches during an overnight polysomnographic sleep study (Sleep. 2019;42(12):zsz180. https://doi.org/10.1093/sleep/zsz180). The analysis showed that the new algorithm had differentiated sleep from wakefulness with an accuracy of 90%, and in the 31 participants studied, it showed higher sensitivity (93%) than specificity (60%). The researchers at the University of Michigan further explained that Apple Watches detected sleep correctly among 93% of the participants but misclassified wakefulness vs sleep.
Potential Benefits and Limitations
Additional clinical trials should ideally be conducted to address the variety of issues that relate to the use of variables, which are on a trajectory to gain even more widespread use. But frequent updating of software will make validation studies difficult if not impossible to perform. That said, there is significant desire for learning more about one’s own physiology—specifically, sleep-wake rhythm—and use of these individualized data may help address specific issues in a specific manner for patients; this precision health may be a plus. Further availability of data on a large number of individuals may help us better understand the physiological variables, range, and pathologies in large populations. Finally, there may be benefit of health care monitoring as well as monitoring for clinical research studies with use of the wearable devices.
Other benefits of using wearable and nearable devices include:
- They can collect data 24/7 in a person’s natural setting compared with polysomnography, which is conducted in the sleep disorder center/laboratory setting and not in a patient’s own bedroom.
- Data can be collected over an extended period.
- Data are collected and available in real time.
- Data can be collected with several different physiological measures, such as movement, heart rate, and respiration. This provides better monitoring than subjective assessment of sleep or sleep diaries based on recollection.
- Population data may be available with the large number of devices in use.
Wearables’ and nearables’ ability to measure sleep quality may have some limitations. Many of them often measure small movements—or lack thereof—to track sleep patterns. With many of these devices, there is not sufficient information available regarding accurate quality and quantity of sleep. As to the quantity, it has been reported that wearables overestimate sleep time in participants. It is understandable, since lack of motion does not always equate to sleep.
Other limitations of using wearable and nearable devices include:
- Most of these devices are not extensively researched or approved by the FDA. While studies are emerging that compare the sleep data obtained from these devices with polysomnography, information is still limited.
- Assessment of quality of sleep vs other traditional approved measures is difficult, because ongoing validation studies are almost impossible since the devices continue to evolve; no sooner does the software algorithm change, new validation has to be done.
- Legal and ethical issues, such as ownership of the data, privacy, and security also need to be considered. While the data obtained via medical devices are covered by the Health Insurance Portability and Accountability Act, data obtained by consumer wearables still have personal health-related information but may not be covered under the same protection as the approved medical devices.
- There may be potential of outside interference as is known to occur with mobile devices.
We are living in an increasingly digital world with a variety of devices around us that can measure not just physiological variables but also our day-to-day behavior. These devices are evolving quickly and are increasingly being used by our patients, by our communities, and by populations at large. We need to be aware of these technologies and try to understand their capabilities as much as possible.
Our patients come back with the data from such devices and will do so even more in the future. They need to know and understand that there may be limitations to the accuracy of the data reported by these devices. Nevertheless, if nothing else, some identified patterns may be helpful, and the devices have brought visibility and focus to the importance of sleep. That I believe, in itself, is a plus. On the other side, there might be too much emphasis on the data obtained from devices, leading to users’ anxiety and concern. Day-to-day variability may lead to daily concerns by users.
Pradeep Sahota, MD, is a professor and the chairman of the Department of Neurology at the University of Missouri School of Medicine and is director of the Sleep Disorders Center and director of the Sleep Medicine Program at University of Missouri Health Care in Columbia, Missouri. He is also a member of the American Neurological Association.