How to Add Sleep Metrics to a Health Research Study

Sleep is one of the most commonly studied and clinically relevant behavioral health domains. It plays a central role in physical recovery, metabolic regulation, cognitive performance, mental health, and overall quality of life. In decentralized and real‑world studies, wearable‑derived sleep metrics offer a scalable way to measure sleep patterns longitudinally, outside of the clinic, with minimal participant burden.

This post explains why sleep metrics are frequently included in health research, when they are appropriate, how they can be configured in a decentralized study using Alethios, and what data researchers receive once a study is live.

Why Include Sleep Metrics in a Health Research Study?

Sleep quality and duration are associated with a wide range of health outcomes, including cardiometabolic risk, immune function, cognitive performance, pain sensitivity, and mood (Buysse, 2014; Walker, 2017). Changes in sleep often precede or accompany changes in other clinical or functional outcomes, making sleep a valuable early signal in many intervention studies.

Wearable‑derived sleep data enables continuous, longitudinal measurement in real‑world settings, complementing both in‑clinic assessments and patient‑reported outcomes. While consumer wearables do not replace polysomnography, multiple validation studies have shown reasonable agreement for sleep duration and timing, and improving performance for sleep staging under certain conditions (de Zambotti et al., 2019).

Sleep metrics should be understood as digital functional endpoints. Their interpretability increases substantially when paired with validated questionnaires, symptom tracking, or physiological measures.

When Sleep Metrics Are a Good Fit

Sleep metrics are commonly used in studies focused on:

• Sleep quality and sleep disturbance
• Recovery and fatigue
• Stress and mental health
• Pain and musculoskeletal conditions
• Metabolic health and GLP‑1–related interventions
• Longevity and aging

They are less appropriate as primary endpoints when sleep is not expected to change over the intervention window, or when participants have highly irregular sleep schedules that are not accounted for analytically.

Supported Devices

Alethios supports wearable‑derived sleep metrics from the following personal device ecosystems:

• Apple Health
• Fitbit
• Garmin
• Google Fit
• Oura
• Whoop

Sleep metrics are standardized for use within Alethios. Participants connect their devices during onboarding, after which sleep data is collected passively according to the study schedule.

How to Add Sleep Metrics to a Study on Alethios

Configuration takes place during the Build phase of the Study Planner.

Step 1: Navigate to Build → Add Digital Endpoint

From the Build section of your study, select Add Digital Endpoint to begin configuring wearable‑derived data collection.

Step 2: Select Digital Endpoints

Within the Digital Endpoints modal, select the sleep metrics you wish to collect. Available metrics include:

• Sleep efficiency
• Seconds in bed
• Seconds asleep
• Sleep stages
• Hypnogram

If enabled for your account, you may also select additional wearable‑derived endpoints at the same time, such as step count, heart rate, or heart rate variability. Multiple endpoints can be collected concurrently from the same connected device, depending on provider support and study design.

Step 3: Configure Participant‑Facing Copy

You will be prompted to define participant‑facing fields that explain what data is collected and why.

Title
Sleep Quality Metrics

Description (optional)
Continuous monitoring of sleep duration, efficiency, and sleep stage distribution to assess changes in sleep quality and recovery in real‑world settings.

Justification
Wearable‑derived sleep data provides objective insight into nightly sleep patterns and recovery, complementing self‑reported sleep quality and symptom measures. This helps researchers understand how an intervention affects sleep behavior over time.

Clear justification improves wearable connection rates. Participants are more likely to connect a device when they understand how sleep data will be used within the study.

Step 4: Choose Providers

Select which wearable providers you wish to support. Participants will see only the connection options relevant to the providers you enable.

Step 5: Set Requirement Rules

If the endpoint is marked Required, participants must connect a wearable device in order to become active in the study. If left optional, sleep data will enrich the dataset without gating participation.

What Sleep Data Do Researchers Receive?

Alethios returns structured, time‑aligned sleep data for each participant, organized by study night rather than calendar day. Data includes:

• Enrollment ID
• Study arm
• Device family
• Sleep session start and end times
• Sleep efficiency
• Wakeup events and out‑of‑bed events, when supported
• Seconds in bed, seconds awake, and seconds asleep
• Sleep stage breakdown (REM, deep, light), when supported
• Hypnogram output, when supported
• Adherence metrics based on data availability

Example Sleep Data Output (De‑identified)

Two design choices can materially improve the interpretability of wearable sleep metrics, especially when sleep staging is an important analytic component.

First, consider adding a short run‑in period. Consumer wearables often show an initial “settling” period where participants adjust to wearing the device and where algorithms may behave differently as they accumulate baseline patterns. A run‑in reduces the risk that early nights distort baseline estimates or create artificial pre/post differences that are actually acclimation effects.

Second, consider enrolling participants who use the same wearable (or restricting to a single device family for a given analysis). Sleep staging, wake detection, and efficiency are derived from proprietary algorithms that vary meaningfully across providers and device generations. Restricting device heterogeneity reduces measurement variance attributable to differing algorithms, improves comparability across participants, and simplifies interpretation in downstream reporting.

This structure allows for longitudinal analysis of sleep patterns, night‑to‑night variability, and alignment with other study endpoints.

Common Uses of Sleep Data

Researchers frequently use wearable‑derived sleep data to:

• Evaluate change from baseline in sleep duration or efficiency
• Correlate sleep with patient‑reported outcomes such as fatigue or pain
• Assess recovery trends over time
• Identify responder subgroups
• Contextualize physiological or biomarker changes

Best Practices

• Pair wearable sleep data with validated sleep questionnaires where appropriate.
• Be explicit about which sleep metrics are primary versus exploratory.
• Account for baseline variability and adaptation to wearing a device.
• Preview the participant onboarding flow to confirm clarity of wearable connection steps.
• Document analytical limitations related to device algorithms and sleep staging accuracy.

Next in This Series

This post is part of a broader Alethios series on wearable‑derived digital endpoints, including step count, heart rate variability, readiness, temperature, and menstrual cycle tracking.

References

Buysse DJ. Sleep health: Can we define it? Does it matter? Sleep. 2014.

Walker MP. Why we sleep: Unlocking the power of sleep and dreams. Scribner. 2017.

de Zambotti M, et al. Wearable sleep technology in clinical and research settings. Sleep Medicine Reviews. 2019.

‍