On December 22, 2021, the Food and Drug Administration (FDA) issued a draft guidance for sponsors, investigators, and other interested parties on using digital health technologies (DHT) to acquire data remotely from participants in clinical investigations. DHTs (such as wearables and sensors) are playing a growing role in clinical research, accelerated by the need for decentralized clinical trials and remote patient monitoring during the COVID-19 pandemic. Though largely directed to study sponsors using DHTs, the guidance raises issues that developers and manufacturers of these technologies will want to be mindful of given the ways in which their products might be used in a broader clinical use case than originally anticipated. Comments on the draft guidance are due March 23, 2022. The full text of the guidance can be accessed here.
What is a DHT?
DHT is defined as a system that uses computing platforms, connectivity, software, and/or sensors, for healthcare and related uses. For example, an app that uses the sensors on a smartwatch to collect heart rate information, may be a DHT. DHTs enable the transmission of data across distances, which increases opportunities for patients to participate in trials and to more frequently collect patient data. The guidance details some key considerations for study sponsors when using DHTs for clinical investigations.
Selecting an Appropriate DHT for the Clinical Investigation
When selecting a DHT for use in a clinical investigation, sponsors should ensure that it is “fit-for-purpose”—meaning that the level of validation associated with the DHT is sufficient to support its use and interpretability in the investigation. In making the determination that a DHT is fit-for-purpose, the FDA recommends that sponsors consider:
- the clinical event or characteristic of disease/condition that is being measured
- the proposed trial population, including whether usability studies are necessary
- design of the clinical investigation
- characteristics of the DHT that may influence trial participant use (e.g., whether a subject would use the DHT for the duration of a trial).
The sponsor must provide an explanation of why the DHT is fit-for-purpose for the clinical investigation in a premarket submission (i.e., an application for 510(k) clearance or approval). The explanation should contain basic information about the DHT (e.g., relevant physical characteristics, information on how it measures the clinical event or characteristics of the disease/condition), a description of usability-related features, and a description of how privacy and security of data collected by the DHT is achieved.
Verification, Validation and Usability of the Technology
Verification and validation studies should demonstrate that the clinical event or characteristic to be assessed (e.g., heart rate) is consistently and accurately measured in the population being studied. FDA provides the example that a DHT that captures steps in a healthy participant may not be applicable for participants with Parkinson’s disease, who may have a shuffling gait. Sponsors should consider the DHT performance specifications, identify conditions under which the DHT functions reliably (e.g., temperature range), and ensure different brands of DHTs (e.g., smartphone apps) will collect the same data.
Validation studies, including usability studies, should assess use of DHTs in the intended population, including comparison of measurements made by the DHT against reference measurements (e.g., steps counted by the DHT versus steps counted by observation), factors that may affect the precision or accuracy of the DHT (e.g., placement of a wearable sensor on a wrist versus a hip), physical interference with measurement, and other factors. FDA notes that sponsors may leverage verification and validation data made available by DHT manufacturers.
Using DHT to Collect Data for Clinical Endpoints
Sponsors using a DHT to collect clinical endpoints must be justified in a premarket application. If a DHT is capturing a novel endpoint, the sponsor should provide information on how the endpoint relates to other well-established endpoints and the reliability of the proposed endpoint. Relatedly, an analysis of the data collected from the DHT should be discussed in a statistical analyses plan which, among other things, addresses how missing or erroneous data will be handled, as well as defines and prespecifies endpoints and per-participant source data.
Unique privacy risks may arise when DHTs are used in a clinical investigation. In particular, the guidance recommends that sponsors consider the risk of potential disclosure of identifiable information via (i) a breach of the DHT and (ii) end-user licensing agreements or terms of service that allow sharing of data with the DHT manufacturer and/or other third parties. Thus, Sponsors should ensure that safeguards are in place designed to secure data collected and transmitted by the DHT. While the guidance also points out considerations for record protection and retention (familiar territory for study sponsors), the use of DHTs may present additional complexities when recording and transmitting this data. For example, the specific data elements and data format collected by the DHT, including all relevant associated metadata, may be different than what a study sponsor has handled with more traditional medical products.
In addition to the above key considerations, the guidance provides a handful of other considerations as well as appendices with specific examples of how different types of DHTs may be incorporated into a clinical investigation. Sponsors using novel DHT technology in clinical studies, or collecting novel endpoints using DHTs, should consider engaging early with the FDA Center responsible for the medical product (i.e., CDER) to discuss use of the DHT, and its specific performance attributes and characteristics, prior to conducting the study. Doing so may save time and valuable resources of a company, should FDA disagree with the proposed DHT or its intended use in the study.