Wearable Devices and Digital Health Technologies (DHTs) in Clinical Trials

FDA’s use of the term “digital health technology” (DHT) is deliberately broader than “wearable” — it covers any system using computing platforms, connectivity, software, and/or sensors for healthcare-related purposes that collects, stores, or transmits data used in a clinical investigation. Wearables (smartwatches, patches, rings, continuous glucose monitors) are the category most people picture, but the regulatory framework applies equally to non-worn sensors placed in a participant’s home, implantable devices, and symptom-tracking mobile apps, whenever the data they generate is intended to support a trial endpoint, safety assessment, or other conclusion submitted to a regulator.

The FDA guidance this term is built on

FDA issued final guidance, Digital Health Technologies for Remote Data Acquisition in Clinical Investigations, on December 22, 2023 (finalizing a draft first issued December 23, 2021). It applies to investigations of drugs, biologics, and medical devices and is aimed at sponsors, investigators, and other stakeholders using DHTs to remotely collect data from trial participants. The guidance addresses, among other things: selecting a DHT that is fit for purpose in a given trial; describing the DHT in regulatory submissions (protocol, statistical analysis plan, and marketing application); verifying and validating the DHT before relying on its output; using DHT-derived data to support a trial endpoint; identifying and managing risks specific to DHT use (connectivity loss, battery failure, participant non-adherence to wear schedule); and data management and retention.

Verification and validation — the core sponsor obligation

The regulatory weight of this guidance sits on one distinction: a consumer wearable is built to be engaging and roughly accurate; a clinical-trial endpoint requires evidence the measurement is fit for the specific clinical question being asked. FDA’s guidance asks sponsors to verify that the DHT’s hardware and software function as intended (technical/analytical performance) and to validate that the derived measurement is clinically meaningful for the concept of interest in the trial’s actual population and use conditions — not just in a general-population validation study run by the device manufacturer. A device validated for step-counting in healthy adults is not automatically validated for gait assessment in a Parkinson’s disease trial population; the sponsor, not the device manufacturer, carries the burden of demonstrating fitness for the specific trial context.

This documentation typically needs to cover: what the device measures and how (sensor type, sampling rate, algorithm), evidence of analytical performance, evidence the resulting endpoint is clinically valid for the trial’s population, the data transmission and storage pathway, and a plan for handling missing or implausible data (e.g., a device not worn for a study day, or a physiologically implausible reading).

How this differs from the broader decentralized clinical trials (DCT) concept

DHT use is often discussed alongside decentralized clinical trials (DCTs), and the two frequently appear in the same protocol, but they are not the same thing. A DCT is an operational model — conducting some or all trial activities away from a traditional investigational site, via local labs, mobile nursing visits, telehealth, and direct-to-participant drug shipment, in addition to or instead of DHTs. A trial can use DHTs for endpoint data collection while still being conducted entirely at traditional sites (a wearable dispensed at an in-person visit, worn at home, with data reviewed at the next in-person visit), and a DCT can exist with no DHT component at all (e.g., local lab draws and telehealth visits with no connected sensor). The overlap is real — continuous remote data acquisition is one of the more common reasons a sponsor decentralizes a trial — but the regulatory questions are distinct: DCT guidance addresses where and how trial activities happen and who performs them; DHT guidance addresses whether a specific measurement tool’s output can be trusted as evidence.

Relationship to Good Clinical Practice

DHT use does not create a separate GCP standard. Data collected via a validated DHT is still subject to the same source-data, data-integrity, and quality expectations set out in ICH E6(R3), including its expanded treatment of technology-enabled and risk-proportionate approaches to quality management. Where a DHT is the source of a trial endpoint, the device-generated record functions as source data in the same sense as a case report form entry, and the same attributability, legibility, and audit-trail expectations apply.

Related concepts

• ePRO (Electronic Patient-Reported Outcomes) — overlaps with app-based DHTs but specifically concerns participant self-report rather than sensor-derived physiological measurement.
• Bring Your Own Device (BYOD) for Research Data Collection — addresses the specific sub-question of whether a participant’s personal device (rather than a sponsor-provisioned one) can serve as the data-collection platform.
• eSource (Electronic Source Data) — the broader category of electronically originated source data that DHT-generated records fall into.
• Decentralized Clinical Trials (DCTs) — the operational trial model DHTs are frequently, but not always, deployed within.