Total Sleep Time as a Digital Biomarker

Total sleep time is the amount of time a person spends asleep during a sleep period, excluding the time spent awake in bed. It is the most basic sleep measure and the one every other sleep metric depends on: sleep efficiency is total sleep time divided by time in bed, and sleep stage percentages are proportions of it. Adults are generally advised to obtain seven or more hours per night, and both short and long sleep durations have been associated with adverse cardiometabolic and mortality outcomes in large cohorts. Measured by questionnaire, sleep duration is subject to substantial recall error, and people with insomnia in particular tend to underestimate how long they sleep. As a digital biomarker, total sleep time is estimated nightly in the participant's own bed, which removes both the recall problem and the first-night effect of a sleep laboratory.

Actigraphy, the long-established research method, classifies each epoch as sleep or wake from wrist accelerometer data and sums the sleep epochs. Consumer devices add photoplethysmography, giving heart rate and heart rate variability, and some add skin temperature, which improves the detection of quiet wakefulness that movement alone misses. Both approaches must first establish the sleep period itself, either from a diary, from a device-inferred bedtime, or from a fixed analysis window, and errors in that boundary propagate directly into the total. The reference standard is polysomnography, which scores sleep from brain activity, eye movement and muscle tone. Against polysomnography, modern consumer devices detect sleep with high sensitivity, and research protocols typically average across at least seven nights because night to night variability is large.

Total sleep time is a standard outcome in insomnia trials and a monitoring measure across psychiatry, neurology, pain and oncology research, where sleep disruption both reflects and worsens symptom burden. It is used in shift work and circadian studies, and as a safety or tolerability signal in trials of drugs that affect sedation or arousal. In the public DiMe endpoint library it is registered across trials in sleep, mental health and urogenital conditions, including as a primary endpoint. Because objective and subjective sleep duration diverge systematically, particularly in insomnia, trials commonly collect both a wearable estimate and a validated questionnaire such as the Insomnia Severity Index, and the gap between them is itself informative.

Wearables tend to overestimate total sleep time, because they infer sleep from stillness and a person lying quietly awake looks like a sleeping person to an accelerometer. This bias is largest in the populations sleep research cares most about, particularly insomnia. Devices also differ in how they define the start and end of the sleep period, which shifts the total independently of any change in the participant. Vendor algorithm updates can move values mid-study, threatening longitudinal comparability, and daytime sleep is often excluded or handled inconsistently. Single nights are unreliable and should not be interpreted in isolation.

• Smith MT, et al. Use of actigraphy for the evaluation of sleep disorders and circadian rhythm sleep-wake disorders: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2018. pubmed.ncbi.nlm.nih.gov
• Chinoy ED, et al. Performance of seven consumer sleep-tracking devices compared with polysomnography. Sleep. 2021. pubmed.ncbi.nlm.nih.gov
• de Zambotti M, et al. Wearable sleep technology in clinical and research settings. Med Sci Sports Exerc. 2019. ncbi.nlm.nih.gov
• Watson NF, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015. ncbi.nlm.nih.gov