Sleep has been recognized by professional medical bodies as essential to health.1 It is recommended that adults should sleep 7-9hrs in the night.2 We are becoming a sleep deprived society.3,4 Chronic sleep deprivation has been associated with daytime sleepiness5, hypertension6, diabetes mellitus7, all-cause mortality8, increased driving accidents9, decreased female fertility10, decreased sperm count11 and negative influence on work performance.12 Also, there are >84 types of sleep disorders listed in International classification of Sleep disorders.13
We as individuals need to take charge of our Sleep as sleep is essential to health and sleep deprivation causes various disease states. Recent studies have shown that people are sacrificing Sleep for sedentary activities14 and pushing back their sleep time for spending more time on smart phones for catching up on social media.15 This has been shown to increase the risk for cardiovascular disease.16
The most accurate way to measure Sleep is Polysomnography (PSG). It’s a technique where a lot of sensors are placed on the patient to record different physiological parameters during the night. It is a whole night procedure where a person needs to sleep for one whole night in the sleep lab. A trained physician or sleep technologist analyses the whole night data next morning to generate a sleep report which tells about persons sleep, respiration, heart activity and body movements. One need to measure following during PSG- electroencephalography (brain activity), electrooculogram (eye movements) and electromyography (muscle activity) to record sleep of the person.
PSG is a very labor-intensive procedure, not readily available, require highly trained manpower, is expensive and time consuming. It is not a good way to measure sleep at population level over multiple nights due to these disadvantages. We need simple technologies that can record sleep for several nights in patients own sleeping environment enabling people to take charge of their sleep and get involved in the management of their overall health.
Ownership of smart phones have increased and a lot of technologies have leveraged this to monitor sleep through apps, gadgets connected to smart phones or through stand-alone devices. This has brought the monitoring of sleep from sleep lab to consumer level.
Consumer sleep technologies (CST’s) refer to computer-based systems available to general public for improving or self-monitoring their sleep. CST’s are not medical grade devices. Their primary goals have been described as Sleep induction, Wake induction, Self-guided sleep assessment, Entertainment, Social connection, Information sharing and Sleep education.
There are five primary delivery platforms for CST’s17–
- Mobile device Platform
- Wearable platform
- Embedded platforms
- Desktop or website platforms
- Accessory appliance platforms
|Mobile device Platform||This consists of mobile apps running on smart phones and tablets. They do not require external sensors except for smart phone or tablet||Sleep cycle Sleep Bot Sleep as android Sunriser Entrain Go! To Sleep||Convenient and easy to use App accessibility Device capability and flexibility||Sleep disruption from noise and light pollution Reduced processing power compared to stand-alone devices Smart phone need to be placed on the bed. Sensor accuracy may suffer due to other people lying on the same bed.|
|Wearable platforms||A sensor is placed directly on the body or attached or embedded in clothing. Track body movement or biometric information||Fitbit Jawbone UP (Discontinued) Smart watches Basis peak Mimo baby monitor Sleep Image Garmin||Increased accuracy with direct contact with wearer||Discomfort Limited battery life Device misplacement Sensor damage Inaccuracy from frequent use|
|Embedded platforms||Non wearable devices embedded in users sleep environment||Tanita sleep scan Sleep number x12 Luna Beddit Dozee||Unobtrusiveness||May raise privacy concerns due to easy concealment|
|Desktop or website platforms||Computer programs or websites designed to run on a full desktop operating system||Medhelp sleep tracker Myapnea.org Somryst (SHUTi) Sleepio SleepyHead||Increased processing power Larger data storage Better exchange of information||Higher cost Decreased portability Large platform variability|
|Accessory appliance platforms||Any physically separate device that may or may not interface with mobile devices or internet||Clocky Philips wake up light emWave Resmed S+ Sense with sleep pill Withings Aura Earlysense live||Feature design flexibility Improved functionality||Increased cost of stand-alone device Diminished space economy|
Table 1: Summary of different CST platforms
Ideally any CST should monitor and report at least the following parameters about sleep-
- Amount of NREM Sleep and its substages- Stage N1, Stage N2 and Stage N3 (Deep Sleep)
- Amount of REM Sleep
- Total sleep time
- Sleep onset latency- It means how long one took to fall asleep after lying down in bed
- Stage REM latency- it means time after sleep onset when first stage REM occurred
- Time at which patient went to bed
- Time at which patient wake up
- Time in bed
- Sleep efficiency
- Wake after sleep onset (WASO)- It refers to time for which patient was awake after sleep onset
But not all currently available CST’s report above parameters. Most of the CST’s report sleep as light sleep, deep sleep and REM sleep rather than telling about specific NREM substages.18 The definition of deep sleep also vary between different manufacturers and it does not always mean Stage N3 of NREM sleep.18
CST should also be able to tell about sleep quality quantitatively19 and predict over all sleep health score20 for consumers to understand about the influence of their overnight sleep on their health. Very few CST report qualitative sleep quality and none report on Sleep health score for meaningful interpretations of overnight sleep data.
Another important parameter to be reported is chronotype of the person. Chronotype means at what time of the day one is most active. Different individuals differ in the time of their peak alertness. Chronotype is of three types-
- Morning type- These people are maximally active during morning time and tend to sleep early
- Evening type- These people are most active in the evening and tend to sleep late
- Neither type
It is very important to understand one’s chronotype to sleep at the right time. An evening chronotype person will find it difficult to sleep between 9-10pm which is usual recommended sleeping time for adults to get 7-9hrs of sleep. I am not aware of a single app that combines chronotype information with above sleep information to provide meaningful interpretation and suggestions for overnight sleep.
CST’s versus PSG
Most of the CST’s use accelerometry to detect body movements and do sleep staging based on this data rather than recording electroencephalography (brain activity), electrooculogram (eye movements) and electromyography (muscle activity) to record sleep of the person like in PSG. When patient is moving device marks it as wakefulness and when lying still the device marks it as sleep. The algorithm to predict light sleep, deep and rem sleep are proprietary and manufacturers do not disclose them to clinicians and researchers which makes it very difficult to do comparative studies comparing different CST’s.
Accelerometry is also used by medical grade technology to record sleep and wakefulness called Actigraphy. Actigraphy is considered mobile sleep assessment standard. It’s a small watch like device worn on non-dominant wrist and reports total sleep time, time in bed, wake after sleep onset, sleep onset latency, sleep efficiency, light intensity and mid-point of sleep. Actigraphy allows for multi night recording of sleep in patients own sleeping environment and is extensively used to record sleep at population level in research studies. But actigraphy overestimates sleep time and underestimation of wakefulness when compared to PSG.
Very few manufacturers provide validation of the sleep data reported by their devices by comparing their device data with PSG or actigraphy. While choosing CST to track your sleep, I will recommend to look for CST’s which report validation of their sleep data against PSG or actigraphy.
A recent study compared seven consumer sleep tracking devices against PSG and Actigraphy in healthy adults.21 They tested four wearable devices (fatigue science readiband, fitbit alta HR, garmin Fenix vivosmart and garmin vivosmart 3) and three non-wearable devices (earlysense live, resmed S+ and sleepscore max) against PSG and Actigraphy. They reported that all these devices except Garmin performed better than actigraphy in detecting sleep vs wake and they tended to perform worse on nights with disturbed or poor sleep.21 We need more studies like this comparing other CST’s with PSG or Actigraphy.
Another study reported on validation of three phone apps (Sleep time, MotionX 24/7, Sleep Cycle) against PSG.22 None of the three apps correlated with PSG and failed to accurately reflect sleep stages.
A recent study reported that only 32.9% of sleep apps contained information supporting their claims, 15.8% included clinical input and 13.2% contained links to sleep literature. Also, apps contained information on how sleep is affected by alcohol or drugs (23.7%), food, daily activities and stress (13.2%). Users gave high rating to apps that contained a sleep tip function.23
One of the issues with validating CST’s data against PSG or Actigraphy is frequent updates of the CST models or their software’s by the manufacturers. By the time a validation study is published, manufacturers either come with a new model with revised algorithms or upgrade the software of the existing devices which makes it impossible to apply validation study results to new models or revised algorithms.
CST across different age groups
Validation of data across different age groups is not available and scant available data shows different results across age groups.
In healthy adults Fitbit Charge 2 when compared to gold standard PSG overestimated total sleep time and time spent in N1+N2 sleep stage, underestimated sleep onset latency and stage N3 but did not differ in estimation of time wake after sleep onset and time spent in REM sleep. Fitbit Charge 2 correctly identified 82% of PSG-defined non-REM-REM sleep cycles across the night in healthy adults and also in subjects with PLMS.24
In pediatric population (3-17yr), Fitbit Ultra significantly overestimated Total Sleep Time(TST) (41 min) and Sleep Efficiency (SE) (8%) in Normal mode, and underestimated TST (105 min) and SE (21%) in Sensitive mode.25
In adolescents, Fitbit Charge HR significantly but negligibly overestimated TST by 8min and SE by 1.8%, and underestimated WASO by 5.6min (p<0.05).26
CST’s and disease management
Scientific community is increasingly interested in using CST’s for treatment follow up of different disease states as it provides multiple night data and allows assessment of the effectiveness of the treatment. Current data in different disease states show different results.
Recently a protocol of a study is published studying the clinical applicability of wearable device (Fitbit Charge HR or Fitbit Charge 2) generated data to the management of thyrotoxicosis by analysing continuously monitored data for heart rate, physical activity, and sleep in patients with thyrotoxicosis during their clinical course after treatment.27
In major depressive disorder patients, Fitbit Flex (FBF) in normal setting significantly overestimated sleep time and efficiency, and displayed poor ability to correctly identify wake epochs. In the sensitive setting, the FBF significantly underestimated sleep time and efficiency relative to PSG.28
Some studies have tested the proof of concept that these technologies can be used as diagnostic tools with varying results. A recent study compared data from Polysomnography with snore data recorded from smartphone taped to patients chest and found good agreement between RDI from smart phone and AHI from PSG.29 However in real life settings the recordings may be affected by the presence of a bed partner and other sounds in sleeping environment including bed partner snoring.29
Data from Sonomat, contactless sleep monitoring system embedded into a foam mattress which detects apneas and hypopneas, was found to have good correlation for AHI, apnea index and hypopnea index for AHI<50events/hr when compared to polysomnography (PSG). 30 Sonomat has recently been validated for detecting sleep disordered breathing in children as well.31
Data from another contact-free monitoring system (EarlySense, Ltd., Israel) was compared to PSG.32 It comprised of an under-the-mattress piezoelectric sensor and a smartphone application, to collect vital signs and analyse sleep. Total Sleep Time estimates with the EarlySense were closely correlated with the PSG. This system also showed good sleep staging capability with improved performance over accelerometer-based apps.32 It can also collect additional physiological information on heart rate and respiratory rate.32
Sleep on Cue, i-phone based app, was compared with PSG to detect sleep onset.33 Sleep on Cue app uses behavioural responses to auditory stimuli to detect sleep onset. Sleep on Cue app overestimated sleep onset latency by 3.17min.33 Total Sleep Time and sleep latency estimated by another i-phone app, Sleep cycle, in Children (2-14yr) did not correlate with PSG.34
Most of the available CST’s are validated in healthy population and need validation in patients with sleep disorders. Studies have shown that CST’s perform poorly in patients with sleep disorders or disturbed sleep.21 My advice is that if you are already diagnosed with sleep disorder or suffer from poor sleep, then see a Sleep physician and do not rely too much on the sleep data shown by CST.
These technologies are becoming an integral part of human lives. Whether we like it or not, patients are coming to the clinic asking to interpret the data of these devices. I usually take a look at the data to see whether it is making any sense clinically. I am happy that these technologies are at least engaging the consumers and making them self-aware and interested in their own Sleep health. But there is a danger that these CSTs may prevent few patients from seeking professional evaluation and treatment or even destroy doctor-patient relationship by providing conflicting advice. Also, app users may be exposed more to their smartphones before bedtime which may have influence on circadian rhythms of the person and make sleep onset difficult.
CST and Insomnia management
The first line of management for Insomnia is cognitive behavioral therapy (CBT). It involves 6-8 1hr sessions with trained Psychologist. CBT is not commonly available and there is a try to offer digital CBT for Insomnia to overcome the problem of trained experts.
CST offer a great chance to achieve the delivery of digital CBT for Insomnia. Clinical studies have shown that digital CBT for Insomnia helps to reduce severity of Insomnia symptoms.35,36 another study have shown that use of digital CBT is effective in improving functional health, psychological well-being, and sleep-related quality of life in people reporting insomnia symptoms.37
A number of digital CBT programs have been developed; the 2 most widely known and fully automated programs are Sleepio and Somryst (previously called SHUTi). Both deliver content and exercises across 6 sessions over a flexible timeline, and content remains available to suit the patient’s needs.
Sleepio is delivered via their website and has an accompanying smartphone application; Somryst was historically delivered via their website and is now available via a smartphone application. Both Sleepio and Somryst use dynamic, user-friendly interfaces to keep patients engaged (eg, animations that illustrate treatment content), and both use patient input (eg, sleep diaries and in-program questions) to personalize the intervention. The US Food and Drug Administration recently cleared Somryst as a prescription digital therapeutic (PDT), which can be prescribed much like pharmacological interventions.
While CST’s have the potential to engage people in their own Sleep health, it has also led to increased preoccupation or concern regarding perfecting the wearable sleep data. This increased concern in perfecting the wearable sleep data has caused sleep problems in some patients and the condition has been termed Orthosomnia.38
I feel that CST’s have a role to play in monitoring Sleep but we need to educate the society about what CST can and cannot do. We need to be clear that CST’s are one of the ways to monitor your sleep health and any abnormal data coming from these devices should alert the person to seek medical consultation rather than keep trying to perfect the data following wrong practices and aggravate the sleep problem.
Which CST should I use to monitor my Sleep?
This is the most common and difficult question asked by my patients when discussing about CST’s. I tell my patients to choose CST by considering the following points-
- First decide what all do they want to measure. Is it sleep only or other things like physical activity as well.
- Do they need information on sleep only or other physiological parameters like heart rate, respiratory rate and heart rate variability as well?
- Look for validation data against PSG or actigraphy
- Check whether data can be shared with the Physician in a meaningful format
- Check whether sleep quality is reported and if reported is it quantitatively evaluated
- Check whether a Sleep health score along with its interpretation is given. Do they explain how sleep health score was computed?
- Check whether sleep tips are included
- Check whether the technology takes into consideration the chronotype information while advising on Sleep
- Check whether you are looking for CST offering treatment for Insomnia
- Check whether the information regarding sleep and solutions for sleep problems are based on scientific literature with appropriate references.
- Check whether the CST alerts you to see a Sleep Physician with a list to choose from
I am sure you will be able to choose the right CST for yourself after going through blog and monitor and track your Sleep for timely interventions when required.
Until then, sleep well and on time.
- Ramar K, Malhotra RK, Carden KA, et al. Sleep is essential to health: an American Academy of Sleep Medicine position statement. Journal of Clinical Sleep Medicine.0(0):jcsm.9476.
2. Hirshkowitz M, Whiton K, Albert SM, et al. National Sleep Foundation’s sleep time duration recommendations: methodology and results summary. Sleep Health. 2015;1(1):40-43.
3. Effect of short sleep duration on daily activities–United States, 2005-2008. MMWR Morb Mortal Wkly Rep. 2011;60(8):239-242.
4. Kocevska D, Lysen TS, Dotinga A, et al. Sleep characteristics across the lifespan in 1.1 million people from the Netherlands, United Kingdom and United States: a systematic review and meta-analysis. Nat Hum Behav. 2021;5(1):113-122.
5. Punjabi NM, Bandeen-Roche K, Young T. Predictors of objective sleep tendency in the general population. Sleep. 2003;26(6):678-683.
6. Gottlieb DJ, Redline S, Nieto FJ, et al. Association of usual sleep duration with hypertension: the Sleep Heart Health Study. Sleep. 2006;29(8):1009-1014.
7. Luyster FS, Strollo PJ, Jr., Zee PC, Walsh JK, Boards of Directors of the American Academy of Sleep M, the Sleep Research S. Sleep: a health imperative. Sleep. 2012;35(6):727-734.
8. Soh AZ, Chee MWL, Yuan JM, Koh WP. Sleep lengthening in late adulthood signals increased risk of mortality. Sleep. 2018;41(3).
9. Stutts JC, Wilkins JW, Scott Osberg J, Vaughn BV. Driver risk factors for sleep-related crashes. Accid Anal Prev. 2003;35(3):321-331.
10. Wang ID, Liu YL, Peng CK, et al. Non-Apnea Sleep Disorder Increases the Risk of Subsequent Female Infertility-A Nationwide Population-Based Cohort Study. Sleep. 2018;41(1).
11. Chen H-G, Sun B, Chen Y-J, et al. Sleep duration and quality in relation to semen quality in healthy men screened as potential sperm donors. Environment International. 2020;135:105368.
12. Appleton SL, Reynolds AC, Gill TK, Melaku YA, Adams R. Waking to use technology at night, and associations with driving and work outcomes: a screenshot of Australian adults. Sleep. 2020;43(8).
13. Sateia MJ. International classification of sleep disorders-third edition: highlights and modifications. Chest. 2014;146(5):1387-1394.
14. Zomers ML, Hulsegge G, van Oostrom SH, Proper KI, Verschuren WMM, Picavet HSJ. Characterizing Adult Sleep Behavior Over 20 Years-The Population-Based Doetinchem Cohort Study. Sleep. 2017;40(7).
15. Chung SJ, An H, Suh S. What do people do before going to bed? A study of bedtime procrastination using time use surveys. Sleep. 2020;43(4).
16. Matricciani L, Dumuid D, Paquet C, et al. Sleep and cardiometabolic health in children and adults: examining sleep as a component of the 24-h day. Sleep Med. 2021;78:63-74.
17. Ko PR, Kientz JA, Choe EK, Kay M, Landis CA, Watson NF. Consumer Sleep Technologies: A Review of the Landscape. J Clin Sleep Med. 2015;11(12):1455-1461.
18. Ong AA, Gillespie MB. Overview of smartphone applications for sleep analysis. World Journal of Otorhinolaryngology-Head and Neck Surgery. 2016;2(1):45-49.
19. Ohayon M, Wickwire EM, Hirshkowitz M, et al. National Sleep Foundation’s sleep quality recommendations: first report. Sleep Health. 2017;3(1):6-19.
20. Brindle RC, Yu L, Buysse DJ, Hall MH. Empirical derivation of cutoff values for the sleep health metric and its relationship to cardiometabolic morbidity: results from the Midlife in the United States (MIDUS) study. Sleep. 2019;42(9).
21. Chinoy ED, Cuellar JA, Huwa KE, et al. Performance of seven consumer sleep-tracking devices compared with polysomnography. Sleep. 2021;44(5).
22. Choi YK, Demiris G, Lin S-Y, et al. Smartphone Applications to Support Sleep Self-Management: Review and Evaluation. Journal of clinical sleep medicine : JCSM : official publication of the American Academy of Sleep Medicine. 2018;14(10):1783-1790.
23. Lee-Tobin PA, Ogeil RP, Savic M, Lubman DI. Rate My Sleep: Examining the Information, Function, and Basis in Empirical Evidence Within Sleep Applications for Mobile Devices. J Clin Sleep Med. 2017;13(11):1349-1354.
24. de Zambotti M, Goldstone A, Claudatos S, Colrain IM, Baker FC. A validation study of Fitbit Charge 2 compared with polysomnography in adults. Chronobiol Int. 2017:1-12.
25. Meltzer LJ, Hiruma LS, Avis K, Montgomery-Downs H, Valentin J. Comparison of a Commercial Accelerometer with Polysomnography and Actigraphy in Children and Adolescents. Sleep. 2015;38(8):1323-1330.
26. de Zambotti M, Baker FC, Willoughby AR, et al. Measures of sleep and cardiac functioning during sleep using a multi-sensory commercially-available wristband in adolescents. Physiol Behav. 2016;158:143-149.
27. Lee JE, Lee DH, Oh TJ, et al. Clinical Feasibility of Continuously Monitored Data for Heart Rate, Physical Activity, and Sleeping by Wearable Activity Trackers in Patients with Thyrotoxicosis: Protocol for a Prospective Longitudinal Observational Study. JMIR research protocols. 2018;7(2):e49.
28. Cook JD, Prairie ML, Plante DT. Utility of the Fitbit Flex to evaluate sleep in major depressive disorder: A comparison against polysomnography and wrist-worn actigraphy. J Affect Disord. 2017;217:299-305.
29. Nakano H, Hirayama K, Sadamitsu Y, et al. Monitoring sound to quantify snoring and sleep apnea severity using a smartphone: proof of concept. J Clin Sleep Med. 2014;10(1):73-78.
30. Norman MB, Middleton S, Erskine O, Middleton PG, Wheatley JR, Sullivan CE. Validation of the Sonomat: a contactless monitoring system used for the diagnosis of sleep disordered breathing. Sleep. 2014;37(9):1477-1487.
31. Norman MB, Pithers SM, Teng AY, Waters KA, Sullivan CE. Validation of the Sonomat Against PSG and Quantitative Measurement of Partial Upper Airway Obstruction in Children With Sleep-Disordered Breathing. Sleep. 2017;40(3).
32. Tal A, Shinar Z, Shaki D, Codish S, Goldbart A. Validation of Contact-Free Sleep Monitoring Device with Comparison to Polysomnography. J Clin Sleep Med. 2017;13(3):517-522.
33. Scott H, Lack L, Lovato N. A pilot study of a novel smartphone application for the estimation of sleep onset. J Sleep Res. 2018;27(1):90-97.
34. Patel P, Kim JY, Brooks LJ. Accuracy of a smartphone application in estimating sleep in children. Sleep Breath. 2017;21(2):505-511.
35. Vedaa Ø, Kallestad H, Scott J, et al. Effects of digital cognitive behavioural therapy for insomnia on insomnia severity: a large-scale randomised controlled trial. Lancet Digit Health. 2020;2(8):e397-e406.
36. Soh HL, Ho RC, Ho CS, Tam WW. Efficacy of digital cognitive behavioural therapy for insomnia: a meta-analysis of randomised controlled trials. Sleep Med. 2020;75:315-325.
37. Espie CA, Emsley R, Kyle SD, et al. Effect of Digital Cognitive Behavioral Therapy for Insomnia on Health, Psychological Well-being, and Sleep-Related Quality of Life: A Randomized Clinical Trial. JAMA Psychiatry. 2019;76(1):21-30.
38. Baron KG, Abbott S, Jao N, Manalo N, Mullen R. Orthosomnia: Are Some Patients Taking the Quantified Self Too Far? J Clin Sleep Med. 2017;13(2):351-354.