Can Consumer Wearable Data Support Clinical Trial Endpoints?

Consumer wearable data can support certain clinical trial endpoints, but accuracy varies by device, physiological measure, and individual participant characteristics. For example, group-level accuracy often masks individual-level variability exceeding an hour for sleep metrics, and unvalidated devices risk diluting real treatment effects. Sponsors should match device selection to the specific endpoint and confirm validation evidence before committing to a data collection strategy.

Key Points 

• • Only about 11% of commercially available wearable devices have been validated against an accepted reference standard for even one physiological outcome, according to a 2024 living systematic review in Sports Medicine.
  • In a suvorexant trial for insomnia in Alzheimer's disease, polysomnography detected a 35-minute treatment effect while a consumer wearable captured only a 20-minute difference for the same participants.
  • Group-level accuracy can conceal individual-level error: one validation study found Oura Ring sleep duration differences under 12 minutes on average, but exceeding plus or minus 60 minutes for individual participants.
  • The FDA's 2023 final guidance on digital health technologies applies a risk-based, fit-for-purpose validation standard to any wearable used to support treatment or safety decisions.
  • The RATE-AF trial captured over 143 million heart rate readings from consumer wearables and found treatment effects equivalent to intermittent ECG monitoring, showing fit-for-purpose validation can succeed for the right endpoint.
  • The full article details manufacturer-specific bias patterns, validation study designs across the full cited evidence base, and a complete FDA and EMA regulatory checklist, content not included on this page 

Why It Matters Now

For any participant enrolled in a trial where a wearable is an outcome measure, the accuracy of that device can meaningfully affect whether a true treatment effect is detected or lost to measurement error. In some cases, that gap has been the difference between a result reaching significance and falling short of it. This is not hypothetical.

A validation study of the Garmin vívosmart® HR in a suvorexant trial for Alzheimer's disease found that polysomnography detected a 35-minute increase in total sleep time, while the consumer wearable measured only a 20-minute difference. More than 4,000 clinical studies registered globally already use wearable technology, and that number continues to rise.

The FDA's final 2023 guidance, Digital Health Technologies for Remote Data Acquisition in Clinical Investigations, sets a risk-based validation standard for any device informing treatment or safety decisions. That expectation is active now, for trials selecting wearable endpoints today.

What Evidence Shows

Clinical teams have had legitimate reasons to treat consumer wearables with caution. Proprietary, closed algorithms restrict independent verification. Mid-study firmware updates can shift measurement outputs without warning. No standardized validation protocol exists across manufacturers, making device comparisons difficult even for experienced statisticians.

A living systematic review found that published validation studies cover only about 3.5% of the device-outcome combinations needed for full evaluation. This analysis draws on that peer-reviewed evidence base together with clinical and product expertise from Signant Health's Science and Medicine team and WeGuide, the wearable connectivity platform, a cross-institutional review intended to separate validated use cases from unproven ones.

For teams currently selecting a wearable device for a Phase 3 endpoint strategy, the full whitepaper details manufacturer-specific bias patterns, the complete FDA and EMA regulatory checklist, and validation study designs across all cited sources that this page does not include.

Authors

Lauren Crooks, MSc, Director, Clinical, Signant Health
Bill Byrom, PhD, Vice President, eCOA Science, Signant Health
Greta Marie van Schoor, PhD, Signant Health
Anthony Everhart, MD, Clinical Vice President, Internal Medicine, Signant Health
Thijs Sondag, Head of Product, WeGuide