INTRODUCTION — Actigraphy is a validated method of objectively measuring sleep parameters and average motor activity over a period of days to weeks using a noninvasive accelerometer [1-6]. The accelerometer is housed in a small device that is worn like a wristwatch.
Actigraphy is more accurate than self-reported sleep duration and, as such, may be more useful than sleep diaries in the assessment of patients with suspected sleep disorders. Because actigraphy is measured in the patient's home environment, it has greater external validity for certain sleep parameters compared with in-laboratory polysomnography.
The main uses of actigraphy are to objectively measure sleep-wake cycles in patients with suspected circadian sleep-wake rhythm disorders and to complement self-reported sleep duration and other sleep parameters in patients with a range of sleep disorders, including insufficient sleep syndrome. Actigraphy can also provide useful follow-up information to assess treatment response.
This topic provides an overview of how actigraphy is used in adults and children with suspected sleep and circadian rhythm disorders and a brief discussion of consumer wearable devices. Clinical features and diagnosis of circadian sleep-wake rhythm disorders and other sleep disorders are presented separately. (See "Overview of circadian sleep-wake rhythm disorders" and "Classification of sleep disorders".)
TECHNIQUE
Preparation
Preparing the device — Actigraphy devices or actigraphs are relatively easy to use and require very little preparation. Most devices use a rechargeable lithium battery that must be charged prior to use. Battery life is influenced by several factors, including the battery type, the epoch length, and special features (eg, light sensor).
The device must be programmed prior to use. This typically involves entering patient information, setting the epoch length, and selecting data collection start and end times. The most validated and commonly used epoch lengths are 30 seconds and 1 minute [7].
Preparing the patient — Patients should be provided with written instructions on use and care of the actigraph. The device is usually worn on the nondominant wrist. Patients should be informed that the actigraph records movement.
An actigraphy log (table 1) should be provided along with an explanation about the importance of completing the log daily for the entire monitoring period [8,9]. Log information improves the accuracy of the data analysis. (See 'Measured variables' below.)
If an event marker will be used, it may be helpful to show the patient a sample of the data obtained by the device, called an actigram (figure 1); it may also be helpful to show how the event marker shows up on the actigram (eg, wake time, bed time, and nap start and end times).
If removed, the actigraphy device should be placed back on the same wrist. Water resistance should be noted and addressed appropriately with the patient. If a light sensor is being used, patients must ensure that clothing does not cover the sensor.
Protocols
General considerations — Actigraphy provides reliable data about sleep in the home environment but does not provide information about sleep architecture or respiration. Actigraphy may have limitations in patients with limited mobility or movement disorders. (See 'Validation and limitations' below.)
In general, actigraphy should be used for two weeks in order to capture both working and nonworking days [10]. The recording period should exclude holidays and travel, unless the purpose of using actigraphy is to assist with frequent jet lag (see 'Occupational populations' below). After the recording period, patients should be asked about any unforeseen events that may have disrupted sleep that were not noted on the actigraphy log (for example, patient or family member illness, acquiring a new pet).
Adult patient populations — Actigraphy has been validated in a variety of adult patient populations [1,2,5,6]. In general, results are most accurate in healthy adults. Even in patients with sleep disorders, which can affect validity of some parameters, actigraphy provides valuable information on sleep-wake patterns, activity, and light exposure. (See 'Validation and limitations' below.)
Actigraphy has been used to evaluate sleep-wake and activity patterns in patients with a broad range of medical and neurologic conditions, including cancer, attention deficit hyperactivity disorder, acquired immunodeficiency syndrome, and traumatic brain injury. Actigraphy has been used successfully in the home environment as well as in nursing homes and inpatient hospital wards.
Occupational populations — Actigraphy is commonly used in the evaluation of circadian sleep-wake rhythm disturbances related to night shift work or frequent international travel. Polysomnography may not be feasible or necessary in such patients and does not capture the full range of symptoms, which may be most prominent in the occupational setting. Actigraphy can be used to monitor changes in sleep on both workdays and non-workdays, as well as during overseas travel.
For suspected shift work disorder, individuals should begin the observational period on the last day of a work shift in order to capture several typical non-workdays as well as the sleep transition back to workday shifts. Shift workers may use different sleep patterns or strategies to switch back and forth from night sleep to day sleep [11,12]. The recording period may need to exceed two weeks for some workers to include a full rotation of shifts. (See "Sleep-wake disturbances in shift workers".)
Child-friendly techniques — Preparation for children is similar to adults, except that parents and caregivers should be included in the discussion regarding the purpose of the actigraph and its care. Specifically, care instructions should include discussion of whether there are periods of play (eg, playing a contact sport) that might result in a damaged or lost device. For younger children, naps should be noted on the actigraphy log.
In infants and children under the age of three years, the actigraph is typically placed on the ankle rather than the nondominant wrist. The majority of validation studies on actigraphy devices and scoring algorithms in young children have been conducted using ankle placement. For children who cannot tolerate wrist or ankle placement (eg, children with autism spectrum disorder), one study has demonstrated the validity of placing the device in a specially designed pocket on the shirt sleeve [13]. Additionally, studies have demonstrated that some settings and scoring algorithms are more accurate in younger populations [14].
Adolescents may need to wear actigraphy for 14 days or more to best capture variability in sleep schedules (eg, weekdays versus weekends) [10]. Parents/caregivers should complete an actigraphy log (table 1) on behalf of a child, although one study suggested that actigraphy data without a log may still have acceptable accuracy [15]. Adolescents should complete their own actigraphy log, as parents/caregivers may not be completely aware of the adolescent's behavior.
INDICATIONS — Actigraphy is a convenient way to objectively measure disrupted sleep patterns and sleep-wake cycles. Actigraphy measures sleep-wake periods over multiple days and nights in the patient's home environment, which is an advantage over single-night, in-laboratory polysomnography in certain circumstances. Actigraphy is not a replacement for polysomnography when electroencephalography is needed to characterize sleep architecture, sleep stage, or abnormal movements during sleep, or when sleep-related breathing disorders are suspected. (See "Overview of polysomnography in infants and children" and "Overview of polysomnography in adults".)
A clinical practice guideline of the American Academy of Sleep Medicine endorses use of actigraphy in the assessment and monitoring of a range of sleep problems in both children and adults, including [5,6]:
●Circadian sleep-wake rhythm disorders – The most common use of actigraphy is in patients with suspected circadian sleep-wake rhythm disorders, such as delayed sleep-wake phase disorder (figure 1) or shift work disorder (figure 2). In such patients, actigraphy data complements self-reported sleep parameters obtained from sleep logs and provides a necessary substitute for self-reported sleep parameters in patients who cannot reliably complete sleep logs. (See "Overview of circadian sleep-wake rhythm disorders", section on 'Clinical manifestations'.)
●Insomnia symptoms – Actigraphy is an important tool in the diagnostic evaluation of insomnia symptoms when there is suspicion for a circadian rhythm sleep-wake disorder as an alternative diagnosis to chronic insomnia. (See "Evaluation and diagnosis of insomnia in adults", section on 'Additional testing' and "Evaluation and diagnosis of insomnia in adults", section on 'Differential diagnosis'.)
●Excessive daytime sleepiness – Actigraphy can be used to verify adequate sleep prior to conducting a multiple sleep latency test when there are concerns about the reliability of self-reported sleep. Similarly, actigraphy may be used to confirm or rule out insufficient sleep in patients presenting with complaints of excessive daytime sleepiness. (See "Quantifying sleepiness", section on 'Multiple sleep latency test (MSLT)' and "Insufficient sleep: Evaluation and management", section on 'Evaluation'.)
●Monitoring response to treatment – Follow-up actigraphy can be a very useful way to objectively measure treatment response and enhance patient education. Follow-up data can be particularly useful for patients undergoing cognitive behavioral therapy, for example, as it provides objective feedback on the impact of newly instituted behavioral changes. Patients with a range of sleep habits and sleep disorders can learn to monitor their own sleep and adopt better daily sleep habits by reviewing actigraphy reports with clinicians.
CONTRAINDICATIONS — There are few contraindications for actigraphy in either children or adults. Rarely, patients may develop local skin irritation or hypersensitivity to materials in the device.
Care should be taken when wearing the device near magnetic resonance imaging (MRI) machines, as MRI compatibility is not well established [7].
INTERPRETING RESULTS
Normative values — Although actigraphy has been validated in a variety of populations against a reference standard (eg, polysomnography) in the same population, there are few studies providing normative data for actigraphically-derived sleep variables among healthy control populations. However, actigraphy-derived sleep values should track closely with age-appropriate clinical recommendations for sleep variables, such as sleep duration (figure 3).
A systematic review of 81 studies of actigraphy provided data on a variety of sleep variables and concluded that actigraphy may be a valuable tool in the assessment of sleep characteristics in patients with a variety of sleep disorders, but that there is a need to establish normative data that account for demographic and developmental factors such as age, sex, ethnicity [5,6]. Two studies focused on pediatric populations have reported normal or reference values derived from actigraphy data [16,17]. Normative values for actigraphy-measured total sleep time are available for individuals six years of age and older in the United States [18].
Measured variables — Clinically validated actigraphs generally come with software that uses an algorithm to score the motion data as "sleep" or "wake." Some software packages have the ability to make this determination without actigraphy log information; however, we recommend that actigraphy log data be entered as a standard practice, as this data improves the accuracy of the analysis.
Specifically, the software will only score immobility as "sleep" during an interval that has been recorded as "rest" in the actigraphy log. A summary of actigraphy log data in the patient report also provides useful biofeedback to patients, as it allows them to see the difference between when they thought sleep was occurring and when they actually slept.
In addition to "sleep" and "wake," the actigraph and log data measure the following variables:
●Average activity counts – Activity counts are measured as an average value over an entire period scored as "wake" by the software algorithm. Average activity counts are provided for each 24-hour period. Prior work has shown that, for wrist-worn actigraphs, motor activity during wakefulness is not very sensitive [19].
●Time in bed – Time in bed is the total amount of time indicated as "rest" according to the sleep log.
●Total sleep time – Total sleep time represents the total number of minutes scored as "sleep" within the 24-hour day.
●Light intensity – Some actigraphs have the ability to measure light intensity, including the three color bands of the visible spectrum (figure 2). This information is useful for determining how light exposure may be contributing to a sleep pattern that is either delayed or advanced. (See "Overview of circadian sleep-wake rhythm disorders", section on 'Functions of the circadian system'.)
●Sleep onset latency – For a given interval of sleep, sleep onset latency is the time elapsed between the beginning of "rest" as noted on the actigraphy log and the beginning of "sleep" as scored by the software algorithm. The validity of this measure relies heavily on the accuracy of the actigraphy log, which can be quite variable from patient to patient [20]. Sleep onset latency measured by actigraphy may therefore be less reliable than sleep onset latency measured by in-laboratory polysomnography.
Derived information
●Sleep efficiency – Sleep efficiency, expressed as a percentage, is the total amount of time scored as "sleep" in an interval that has been denoted as "rest."
●Diurnal variation
•Intradaily variability – Intradaily variability quantifies the variation in activity levels within one day.
•Interdaily stability – Interdaily variation quantifies the variation in activity levels from day to day.
●Number of sleep periods – The number of sleep periods is the total number of events scored by the algorithm as "sleep" during an interval that has been denoted as "rest."
●Wake after sleep onset – Wake after sleep onset is the total number of events scored by the algorithm as "wake" during an interval that has been denoted as "rest."
●Fragmentation index – During periods scored as "sleep," the fragmentation index is the total amount of time that the person is mobile divided by the amount of time the person is immobile (continuous nonmoving period). This measure is also referred to as the index of restlessness and is an indication of restless sleep.
●Midsleep time – Midsleep time is the midpoint between sleep start and end times. It serves as an indicator of circadian phase.
Actigraph data review — Qualitative review of the actigram is just as important as review of measured variables and derived information. The indication for ordering the test determines what the clinician should specifically look for in the actigraph data.
Data downloads show the 24-hour time interval on the X axis and sequential days on the Y axis. Motion activity registers as a signal: high signal activity is associated with wakefulness, while the absence of signal activity is associated with sleep (figure 1).
First, clinicians should ensure that the patient has worn the actigraph and that there is adequate monitoring time. If not, patients should receive appropriate education and the test should be repeated.
For all patients, the actigram should be examined to determine if there is a regular sleep-wake pattern over the entire recording period. Other information to examine includes:
●Is there more than one sleep period over a 24-hour day?
●Are naps taken on a regular basis?
●Are naps associated with delayed sleep onset on nap days?
●For each 24-hour day, what is the total sleep time, and is it adequate?
●Does sleep duration and timing differ between work days and off days?
Actigraphy assists in the evaluation of circadian sleep-wake rhythm disorders, and each of these disorders have characteristic findings on actigraphy. Mathematical models are available to estimate dim light melatonin onset from wrist actigraphy data in both non-shift and shift-working adults [21].
●Delayed sleep-wake phase disorder shows a habitual delay in the sleep onset of the major sleep episode (eg, 4:00 AM) and a delay in awakening (eg, noon) (figure 1). (See "Delayed sleep-wake phase disorder".)
●Advanced sleep-wake phase disorder shows a habitual advancement or early onset of the major sleep episode (eg, 6:00 PM) and early awakening (eg, 3:00 AM). (See "Advanced sleep-wake phase disorder".)
●Irregular sleep-wake disorder is characterized by recurrent irregular sleep and wake periods (at least three) throughout the 24-hour day (figure 4). There is not one major sleep period, but multiple sleep periods at irregular times throughout the 24-hour day. There is day-to-day variability without a predictable major sleep-wake pattern. (See "Sleep-wake disturbances and sleep disorders in patients with dementia", section on 'Difficulty falling or staying asleep'.)
●In patients with non-24-hour sleep-wake rhythm disorder, actigraphy shows a pattern of sleep and wake times that are delayed each day by a few minutes to as much as an hour daily, depending on the patient's circadian period. For example, if a patient's endogenous circadian period is 24 hours and 20 minutes, the actigram may show a delay in both sleep onset and awakening by approximately 20 minutes daily.
●In shift work disorder, actigrams may show different sleep and wake times depending on work schedule, short sleep durations, and naps (figure 2).
Patients with insomnia may spend too much time in bed when they are not sleeping or may underestimate their total sleep time. Actigraphy can illustrate these behaviors and provide an objective way to measure treatment response.
VALIDATION AND LIMITATIONS — Actigraphy has been validated in a variety of populations [1,3,5,6]. Compared with a gold standard of polysomnography, actigraphy is very accurate for identifying periods of sleep but is less accurate for identifying sleep onset and periods of wakefulness during sleep. When actigraphy is compared with polysomnography, its accuracy is approximately 90 percent for total sleep time but only 55 percent for determining the correct sleep stage [7].
Total sleep time and sleep efficiency tend to be overestimated by actigraphy, primarily because the delineation of sleep onset is difficult. This leads to overestimation of sleep time in situations in which patients lie in bed relatively motionless (eg, patients with insomnia, those who lie in bed watching television, older adults in a nursing home environment). In contrast, actigraphy may underestimate sleep in patients with a movement disorder.
Another limitation compared with polysomnography is that actigraphy is not able to identify stages of sleep. (See "Overview of polysomnography in adults", section on 'Measured variables' and "Stages and architecture of normal sleep".)
CONSUMER WEARABLE DEVICES
Validation and interpretation — A large number of consumer wearable and nearable devices purport to assess sleep-wake periods, total sleep time, and even sleep stages. However, devices may not use an accelerometer validated for identification of sleep-wake periods. In addition, the algorithms used to determine sleep-wake periods are proprietary and often unavailable to clinicians and the public. As such, few devices have been validated for measuring sleep-wake periods.
Although the data are not entirely consistent, most studies have found that consumer wearable devices overestimate total sleep time and sleep efficiency and underestimate wake after sleep onset compared with polysomnography [22-27]. Nevertheless, technology improves over time, and manufacturers' proprietary algorithms are constantly being updated. Patients are increasingly using these devices and looking to clinicians for guidance about what the data may mean and how to improve their sleep.
Some devices use photoplethysmography, a technology that monitors blood flow in tissue over time via flashing green light-emitting diodes (LEDs) that sense heart rate. Blood absorbs green light, and the higher the blood volume, the more green light is absorbed. This absorption amount can be quantified over time to show a "pulse" rate. Some devices use a combination of movement data and heart rate variability data, which fluctuate and correlate with wakefulness and the different sleep stages [28]. Manufacturers' applications can integrate the data to display sleep information, including sleep stages. Applications may also include benchmark data for what is considered normal and provide educational material about sleep and ways to improve sleep.
All of these data should be interpreted with caution. Devices rely on indirect measures of sleep, have data loss problems, and some devices are more accurate than others. The more advanced devices correlate approximately 70 percent of the time when compared with actigraphy and polysomnography [29]. Some devices try to parse out "light sleep" and "deep sleep," although the terms are not standardized and there is little validation on the accuracy of these designations. One small study in healthy adolescents found that a consumer device accurately estimated sleep duration but not staging compared with clinical-grade actigraphy and polysomnography [30].
Advice for clinical use — Position statements and technology summaries are available from the American Academy of Sleep Medicine (AASM) [31-33]. The AASM encourages clinicians to become familiar with the disadvantages and potential benefits of consumer sleep technologies, including wearable and nearable devices [31,33].
While the devices are not validated and cannot be used for diagnosis or treatment of sleep disorders, patient-generated data may enhance the clinician-patient interaction when reviewed in the context of a clinical evaluation. Use of a device may indicate a patient's commitment to focus on sleep wellness, and the data may provide an opportunity to educate patients about the importance of healthy sleep habits. In some cases, data may point towards the presence of a sleep disorder and prompt formal sleep testing.
When reviewing wearable device data with patients, we suggest the following approach:
●Review variables that define healthy sleep: duration (seven to nine hours), sleep timing (consistent sleep-wake schedule), sleep quality, and the absence of sleep disorders [34]. (See "Insufficient sleep: Definition, epidemiology, and adverse outcomes", section on 'How much sleep do we need?'.)
●Ask patients if the data correspond with their subjective sleep experience. Educate patients that the device may not be accurate and that it is best used as a tool to set personal goals and assess change over time.
●Review good sleep hygiene when appropriate and suggest that patients monitor the device data in response to behavioral changes. Multiple interruptions in sleep may reflect a sleep disorder or an environmental issue (eg, a dog in the bed waking the patient up). (See "Cognitive behavioral therapy for insomnia in adults", section on 'Sleep hygiene' and "Insufficient sleep: Evaluation and management", section on 'Management'.)
●Some device manufacturers have useful sleep-related information on their website or applications. In some cases, sleep specialists have written and reviewed this information.
●Be aware that use of the device may aggravate anxiety, unreasonable sleep expectations, and inadequate sleep hygiene in some patients.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Parasomnias, hypersomnias, and circadian rhythm disorders".)
SUMMARY AND RECOMMENDATIONS
●Definition – Actigraphy is a validated and reliable method for measuring sleep phase, duration, daily patterns, and efficiency over multiple days. The actigraphy device contains an accelerometer and is typically worn on the nondominant wrist. (See 'Technique' above.)
●Indications for use – The most common use of actigraphy is in patients with suspected circadian sleep-wake phase rhythm disorders such as delayed sleep-wake disorder or shift work disorder. Actigraphy is also used to complement self-reported sleep duration and other sleep parameters in patients with a range of suspected sleep disorders and to document response to treatment. (See 'Protocols' above and 'Indications' above.)
●Measured variables – When combined with an actigraphy log (table 1), variables assessed by actigraphy include average activity during waking, sleep efficiency, sleep duration, sleep phase (eg, sleep onset, midsleep time, and wakeup time), variability of sleep, and light exposure. (See 'Measured variables' above and 'Derived information' above.)
●Validation – Actigraphy has been validated against polysomnography and provides accurate estimates of sleep patterns in a variety of patient populations. Total sleep time and sleep efficiency tend to be overestimated by actigraphy, primarily because the delineation of sleep onset is difficult. (See 'Validation and limitations' above.)
●Consumer wearables – Data from consumer wearable and nearable devices that assess sleep should be interpreted with caution. Devices rely on indirect measures of sleep, have data loss problems, and some devices are more accurate than others. Consumer devices are not validated to be a screening tool for sleep disorders. (See 'Consumer wearable devices' above.)
Nonetheless, patient-generated data may enhance the clinician-patient interaction when reviewed in the context of a clinical evaluation. Use of a device may indicate a patient's commitment to focus on sleep wellness, and the data may provide an opportunity to educate patients about the importance of healthy sleep habits. (See 'Advice for clinical use' above.)
