The FDA’s four medical product centers have jointly published a framework for how digitally derived measures should be developed and used in clinical trials. The August 2026 paper covers wearables, sensors and AI-enabled platforms that feed trial endpoints.
The document unifies scattered guidance into a single evidentiary pathway with three stages. Verification confirms the device accurately measures the underlying physical or chemical parameter. Analytical validation shows the device captures the clinical event in the target population. Clinical validation proves the derived measure reflects a meaningful aspect of health and tracks relevant change.
A tiered evidence standard is calibrated to regulatory context. A digitally derived measure serving as a primary endpoint in a pivotal trial faces a higher bar than one used for exploratory analysis or patient engagement.
The paper is not binding guidance. It signals current agency thinking and gives sponsors a roadmap for when wearable data can support a submission. That clarity matters: digital health technology vendors and trial sponsors have long complained that FDA expectations vary between divisions, and the cross-center sign-off is the agency’s answer.
The four centers behind the paper are CDER, CBER, CDRH and the Oncology Center of Excellence, which co-signed to show the framework applies across drugs, biologics, devices and oncology trials.
For AI-enabled platforms, the framework settles a recurring question: a model that turns sensor data into a trial endpoint must be validated on both the measurement and the clinical claim. Sponsors that plan for the three-stage path early should face fewer surprises at submission time.
