Staying current with actigraphy devices for sleep-wake monitoring

By Makayla Cordoza, PhD, RN; Steven Holfinger, MD, MS; Shalini Paruthi, MD, FAASM; Sachin Shah, MD, MS; Cameron Barber, DO; Evin Jerkins, DO; Jennifer Y. So, MD; Navin Verna, MD, FAASM; Sharon Schutte-Rodin, MD, FAASM; on behalf of the AASM Emerging Technology Committee

With the public’s growing interest in sleep and activity habits, the market for actigraphy-based wearable devices designed to track movement and estimate sleep has rapidly expanded. For 2024, approximately 560 million wearable unit shipments were expected, reflecting their increasing relevance beyond sleep clinics to homes and the workplace. Actigraphy devices have evolved significantly in recent years, incorporating additional sensors such as photoplethysmography (PPG) alongside traditional triaxial accelerometers. Wearables typically worn on the wrist are also now available for the finger, arm and chest, offering extended monitoring of activity and sleep. The additional incorporation of artificial intelligence and multi-sensor algorithms to process and interpret wearable data for both consumers and clinicians has also enhanced the potential usefulness of these devices.

This article provides a broad overview of clinical applications for actigraphy-based wearables designed to be worn continuously (both day and night), including technical considerations and a summary of some currently available devices. The AASM and authors have no affiliation with any products and do not endorse any products described here.

Clinical devices for actigraphy in sleep medicine

Actigraphy is an essential tool in sleep medicine, particularly for objectively evaluating sleep-wake patterns and aiding in the diagnosis and the treatment monitoring of sleep disorders. Common indications for actigraphy include its use prior to the Multiple Sleep Latency Test (MSLT) when evaluating central disorders of hypersomnolence to ensure insufficient sleep and circadian misalignment do not confound the results. Actigraphy is frequently used to assess circadian rhythm sleep-wake disorders and to evaluate insomnia characteristics. Actigraphy may also be employed to monitor the effectiveness of treatment interventions, such as response to cognitive-behavioral therapy for insomnia (CBT-I) or pharmacological treatments.

Selecting an appropriate actigraphy device involves several considerations

• Added features: Examples include temperature, PPG, gyroscope, microphone, the ability to wear on different body locations, and other additional features that may be of interest when choosing a device.
• Data outputs and reports: Some wearable companies offer the ability to score raw data, use open-source algorithms, create custom reports and more.
• Remote monitoring features: Ability to view data in real time (to provide immediate feedback and monitor compliance) vs downloading data upon device return.
• Battery life: Minimize interruptions for required charging during monitoring.
• Device validation: Not all devices have validation studies against polysomnography that confirm accuracy and reliability for sleep indices. Considerations include: (1) published performance studies, especially what subjects and data outputs the validation included (e.g., adolescents vs adults, sleep duration, sleep staging, etc.), (2) when FDA-clearance is needed: use indications, inclusion/exclusions and predicate device comparisons, and (3) upcoming and ongoing studies, subjects and results from clinicaltrials.gov.
• Special populations: Accurate data capture requires wearable devices to have an adequate fit to optimize usage. Sizing limitations are a consideration for over/underweight and pediatric populations. The usability of the device (e.g., screen size, the need to navigate buttons, etc.) is also important and population dependent.

Actigraphy performance and validation in pediatric populations present unique challenges. Children’s activity patterns and sleep architecture differ from those of adults, necessitating tailored algorithms for accurate analysis. Studies comparing actigraphy to polysomnography in children generally demonstrate agreement in measuring total sleep time and sleep efficiency. However, interpreting actigraphy movement after sleep onset as wake or sleep, especially in children with movement disorders or developmental delays, presents performance challenges. Further research is needed to enhance algorithm precision for pediatric populations and to ensure reliable results in diverse clinical contexts.

Billing and reimbursement for remote patient monitoring (RPM) with actigraphy depends on proper coding and compliance with insurance requirements. Commonly used billing codes include CPT 99453, 99454 and 99457, which cover device setup, data transmission and patient interaction. Insurance coverage often requires the use of FDA-cleared devices and documented medical necessity. Reimbursement rates vary by state and insurer but are frequently available for pre-MSLT assessments, circadian rhythm evaluations, and pediatric monitoring when supported by adequate documentation. Medicaid and private insurers may cover these services, particularly if they align with telehealth and RPM policies.

Consumer wearables

Competition is high for consumer sleep devices, leading to both rapid technological advancements and shortcomings in available data to judge device performance. Consumers have benefited from longer battery life, improved comfort, intuitive app interfaces and a recent influx of smartwatches/bands, rings and other wearables. Consumer sleep tracking falls under the health and wellness category, which does not require FDA oversight. Consumers rely on device outputs without much demand for performance metrics. For example, many devices offer sleep staging with little validation of their accuracy outside of laboratory settings. Most manufacturers protect analytic algorithms, meaning access to raw data and population datasets are limited. Multiple publications have found consumer device accuracy varies by device and manufacturer. When evaluating consumer device performance for clinical applications it is important to consider the device model and software version. For example, device performance from the same manufacturer may vary, and software updates may alter accuracy.

Several consumer companies are moving into the health care space and re-marketing their devices for medical monitoring. For example, Happy Ring was originally advertised as a consumer device for mood and stress monitoring. It recently rebranded as Happy Health with the FDA-cleared Happy Ring for physiologic monitoring, including sleep, as part of its at-home health monitoring platform. Other consumer-available actigraphy-embedded devices have received FDA clearance for certain physiologic monitoring (e.g., Masimo W1) and for sleep indications (e.g., Apple Watch and Samsung Galaxy Watch for sleep apnea detection).

Sensor analysis

Actigraphy-based wearables use algorithms to identify aspects of sleep by using (at least) motion data from an accelerometer. A common issue with accelerometry-based algorithms is the overestimation of sleep if the user is sedentary or lays still in bed. Algorithms are also predominantly validated for nighttime sleep and are less reliable for detecting daytime sleep.

Fundamentally, an accelerometer identifies spatial displacement, which is the basis for sleep-wake estimation. Additional sensors like ambient light, body temperature or PPG (which can be used to report heart rate variability, respiratory rate and SpO2) can be used to improve the sleep algorithm and provide more data outputs. These parameters combined with microphone analysis create the potential for detection of additional sleep disorders.

Summary

Actigraphy can be a useful clinical tool for remote and long-term sleep monitoring and for assisting in the diagnosis and treatment efficacy for many sleep disorders. Advancements in wearable technologies have revolutionized both consumer and clinical applications related to assessing aspects of sleep and wake behaviors, sleep disorders and monitoring treatments. There are now numerous options for both consumer and medical actigraphy, with ongoing integration of added sensors, updated algorithms/AI, and other capabilities. These advancements come with benefits and challenges for clinicians and researchers in both choosing a technology for use, data management, costs, and in interpreting results for patients. Importantly, unclear and limited insurance reimbursement for actigraphy currently limits routine integration in clinical sleep and circadian care.

With the interest in sleep health monitoring and FDA clearance from consumer wearables growing, an ongoing increase in the number of patients presenting with findings from their apps, wearables and nearables is likely. Staying current with these rapidly evolving consumer and clinical sleep tracking technologies, as well as implementing clinical practice integration, remains challenging. Related AASM member resources include #SleepTechnology, Montage magazine, Talking Sleep, the AASM Engage members forum, publications (e.g., JCSM), courses (e.g., Sleep Medicine Trends, Sleep Medicine Disruptors and the SLEEP annual meeting), and many website links.

Table 1: Examples of commercial actigraphy devices intended to be worn both day and night that monitor sleep/wake patterns

Table 2: Examples of clinical and research-focused actigraphy devices and their features

* All devices use triaxial accelerometers
**All provide access to raw data unless otherwise stated
Abbreviations: Photoplethysmography (PPG), electrodermal activity (EDA), inertial measurement unit (IMU), psychomotor vigilance test (PVT)