Technology of sleep monitoring… consumer beware!

In contrast to gold standard medical sleep monitoring, which entails a full night of monitoring in a lab and hours of specialist analysis, commercial sleep monitoring is portable, accessible and affordable, often consisting of motion sensors worn on the wrist or even an app on a smartphone. But are these products actually effective in monitoring our sleep? Here to explore this issue is sleep expert Dr. Magdy Younes.

Most people have occasional bad nights and many regularly have bad sleep, feel unrefreshed in the morning or feel groggy during the day. Over the past 10-15 years scientific data has linked bad sleep to serious medical disorders including hypertension/cardiovascular disease, diabetes, mood disorders, overall mortality and even obesity.

In parallel, health consequences of poor sleep were widely reported in news media. Suddenly, poor sleep became a major public concern and millions of people wanted to know if they have a sleep problem. This explosive popular demand presented a great opportunity for consumer device manufacturers, and they dutifully responded. Dozens of consumer products that claim to monitor/improve sleep became available and cost as little as $20.

Medical-Grade Sleep Monitoring

The patient spends a full-night being monitored by >20 sensors that record brain activity, various aspects of breathing, electrocardiogram, limb movements and body position.

The gold standard for sleep evaluation, the in-laboratory polysomnogram, is vastly different to evaluation by consumer products. Here, the patient spends a full-night being monitored by >20 sensors that record brain activity, various aspects of breathing, electrocardiogram, limb movements and body position. The recording is continuously supervised by technologists who can identify and fix technical problems thereby ensuring an adequate study.

The extensive recorded data is reviewed by a highly-trained technologist, a process that takes 1-3 hours, to determine how much time was spent awake and in light, moderate, deep and in REM sleep, number of brief arousals and frank awakenings, presence, number and type of breathing abnormalities, and number of abnormal movements. Body movements, the main monitored parameter in consumer products, play no role in determining quality of sleep in these in-laboratory tests.

How the Medical-Device Industry Responded to Increasing Demand

Initially, the medical-device industry introduced portable versions of the in-laboratory systems for home use. These still required specialized technologists to apply the sensors and score the data. There was little reduction in cost or wait times.

This was followed by several medical-grade systems with limited signals designed specifically for home diagnosis of sleep apnea (Home Sleep Apnea Tests). These do not require expert technologists to set up at home or to score. However, without brain monitoring, the impact of sleep apnea on sleep quality is not determined. This, along with a high failure/inconclusive rate, has limited the use of these monitors to patients with high pre-test probability of moderate/severe sleep apnea.

Based on the idea that people tend to move less when asleep, monitoring of body movement (actigraphy) was evaluated as a cheap, convenient way to monitor sleep quality. Validation studies showed that in normal subjects and insomniacs, actigraphy missed, on average, two-thirds of awake time because many patients stay motionless while awake.

By contrast in some disorders with excessive motion during asleep (e.g. sleep apnea and periodic-limb-movements) actigraphy may underestimate sleep time by  up to 200 minutes . Because of these problems, use of actigraphy for evaluating sleep quality in clinical sleep medicine is limited to specific applications.

Finally, with rapid advances in microprocessors and in digital scoring of the EEG at least three small portable brain monitors that can be applied by the patient, with the collected data scored digitally, have recently become available. These have the potential of providing high quality, affordable and comprehensive sleep evaluation at home.

Consumer Sleep Monitors

Anything goes so long as the label states that the device is not intended for medical use.

Encouraging early reports on the use of actigraphy, simplicity of motion detection technology, and lack of regulatory requirement to validate claims made by manufacturers led to proliferation of consumer sleep monitors. Anything goes so long as the label states that the device is not intended for medical use.

Literally hundreds of such devices became available, mostly consisting of motion detectors worn around the wrist  or within sensors or smartphones placed under or on top of the mattress. One device employs remote radio-waves to detect motion and breathing pattern.

How Accurate are Consumer Sleep Monitors?

Of the multitudes of these devices only a handful (n=7 devices) has been subjected to proper validation studies against polysomnography. Of these, four were done on healthy, mostly young, adults, two on healthy children and adolescents, two on insomnia patients, and one each in adults with depression, children with sleep apnea and adults with sleep apnea.

More research is clearly needed to establish validity of other devices, tested under other circumstances, and in other disorders. However, even with the limited information available certain conclusions can be made. More details can be found in two recent reviews.

  • As with actigraphy, in healthy adults and adults with insomnia or depression, consumer devices overestimate sleep time and miss approximately two-thirds of awake time.
  • In healthy children and adolescents, overestimation of sleep time and missing awake time are not as prominent. Interestingly, these systematic errors begin to appear around age 13 and progressively increase as age increases. This is likely because, alas, children do not spend much time being quiet when awake at night.
  • Even when average results of sleep/awake times are close to polysomnography data, individual results can differ from actual individual values by extensive amounts.
  • There is no convincing support for the claim that these devices distinguish light from deep sleep or REM from non-REM sleep.

Implications for the Individual Consumer

Consumer products do have a number of positives: they provide useful information in population-wide tracking of bedtime/wake-time patterns, they serve to increase awareness of sleep issues among users and their physicians, and they demonstrate society’s urgent desire to understand and improve sleep – needs that the medical community must address. However, at present, their value as a screening tool for assessing sleep quality in a single individual is questionable.

Because of their tendency to underestimate awake time, these devices may lead to false reassurance when a sleep disorder exists.

The consumer should not be re-assured by claims that the device has an average agreement of 80% or 90% with polysomnography. What is of relevance to the individual consumer is how much difference may exist between his/her result and the truth. Available data indicate that these differences are sufficiently large that no meaningful information can be obtained with respect to an individual’s sleep quality.

Because of their tendency to underestimate awake time, these devices may lead to false reassurance when a sleep disorder exists. Conversely, excessive awake time may be scored where no disease exists; many people have benign body movements during sleep or stay in bed longer than their sleep requirement (not everybody needs 7 hours of sleep).

In summary, until scientific data becomes available to recommend specific devices for do-it-yourself screening for sleep disorders, it is advisable not to rely on the currently available consumer products. In the meantime, one should seek medical advice if he/she is frequently unrefreshed in the morning or feels sleepy during the day, or if there are indications of disturbances during sleep, such as messy bedcovers in the morning or bedpartner reporting habitual snoring particularly with breathing interruptions, excessive/unusual movement or kicking…etc.

Several consumer products provide feedback that is claimed to improve sleep. Validity of these claims is dependent on accuracy in assessing sleep quality and,  evidence that the results are consistent from night to night. To my knowledge, no such evidence exists. The change in sleep time with repeated use when following device guidance must exceed the changes that occur without guidance before such claims can be made. Furthermore, there is no scientific evidence that, in the absence of any symptoms, reducing awake time by several minutes when it is already in the normal range (e.g. <40 minutes) is beneficial to health.

Why not check out our other guest blog by Dr. Magdy Younes on sleep analysis waking up to the digital age.


View the latest posts on the On Health homepage