The careful selection of sensor technology prioritized in clinical trials, considering regulatory compliance, data security, device characteristics, and the specific needs of the trials for successful and confident implementation.
The past decade has seen increased adoption of smartphones worldwide, paired with a proliferation of biosensors for a wide range of applications from biometric measurements to early diagnostics.
Expansion is set to continue, with 70% of clinical trials expected to incorporate sensors by 2025.1 This includes medical- and consumer-grade sensors, with both types being built with the ability to monitor continuously and communicate data in real-time.
The increasing number of products creates a vast amount of choices in healthcare, including clinical trials. But even with the market growth, the use of sensor technology applicable to clinical trials is still in its infancy. Guidance from governing organizations and regulators is slowly developing, but there is a lack of alignment on the criteria for optimal sensor selection among the multiple stakeholders in clinical trials.
Amid the choices and pathways of emerging technologies, we highlight the differences in characteristics, dependencies, and implications between medical- and consumer-grade products. These factors are critical to consider for a well-designed sensor solution that will facilitate successful implementation within a clinical trial.
First, what makes a device a medical-grade device?
Medical-grade devices often meet the regulatory definition of a device. Section 201(h) of the FD&C Act (21 USC 321[h]) provides that the term “device” means:
- A product that is intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease.
In the United States, the FDA oversees the regulation of devices and in the European Union, the European Medicines Agency (EMA) oversees regulation via the Conformité Européenne (CE) Mark. Other notable organizations that oversee regulations include Health Canada and the Pharmaceuticals and Medical-grade Devices Agency (PMDA) Japan.
Medical-grade devices are classified based on several factors, including the health risk posed to patients. Per FDA and EU regulations, devices are labeled with the following classifications:
On the other hand, consumer-grade devices are intended for everyday use, typically in private homes. Some have received FDA clearance for certain elements of their product, whilst other physiological measures via the same device are marketed as a “general wellness” feature and therefore do not need FDA approval.2
As the market and options for sensors grows, we expect to see an increase in FDA cleared elements within both medical-grade as well as consumer-grade devices.
Ten key considerations for sensor assessment and selection within clinical trials
1. Data security and privacy
Medical-grade device manufacturers comply with regulatory requirements for privacy and security, such as FDA 21 CFR Part 11, Health Insurance Portability and Accountability Act of 1996 (HIPAA), and EU GMP Annex 11 regulations for their data ingestion and data visualization platforms, to the extent possible and intended.
Compliance with regulations includes confirmation of participant consent to ingest and display de-identified data to third parties. Compliance also means that the data acquisition, storage, and handling is designed with clinical trials in mind.
Consumer-grade device companies are not required to conform to global clinical trial regulations for electronic records such as 21 CFR Part 11. This implies that audit trails are not available and any changes to the source data are not traceable back to the origin of the change. The consumer-grade device company could also retain a copy of the original and source data- thus rendering the risk of non-compliant processes in source data handling.
2. Device selection based on endpoint degree
Digital health technologies (DHT) use computing platforms, connectivity, software, and sensors for health care and related uses.3 The FDA guidance includes recommendations for selection of DHTs, including sensors, based on suitability for use in the clinical investigation.4
Recommendations for medical-grade vs. consumer-grade devices based on endpoint degree are rare. It might be appropriate to use commercial devices if study endpoints are exploratory, or for cases in which there are no medical-grade devices available, such as a smartphone camera-based application that can derive physiological measures. Consumer-grade devices may also be applicable if the sponsor specifically wants to compare consumer-grade and medical-grade devices, or if a medical-grade device’s measures have not been validated in the applicable therapeutic area.
3. Display of sensors data for a clinical trial
The display of digital biomarkers from sensors utilized in clinical trials is controlled to minimize potential bias. Surfacing sensor data, such as blood glucose or number of steps, is often considered within medical-grade app development, given the inclusion in clinical trials. There is often functionality to toggle from either “view” or “hide” (aka “blind”) the data from participant only or both participant and study team.
In contrast, most consumer-grade mobile apps share the heath data derived from the sensor with their end users as part of default design. When applied to clinical trials, the visibility into participant health data can create bias and therefore, undesirable influence into trial outcomes.
4. Technical characteristics
The physical design of the sensor such as size, location of device on body, whether device is waterproof, durability of straps, battery life, charging requirements, and impact on daily life should be considered. Most consumer-grade devices are designed with style and form in mind and typically only require pairing with mobile applications to transmit data, as they are marketed to appeal to a wider cohort of users.
Whilst style may not be topmost in mind with medical-grade device companies, they prioritize patient centricity and accuracy of measurements. Certain medical-grade devices may require a data hub/cellular connectivity for data transmission. If the participant will be traveling, then the data transmission mechanism should be considered.
5. Firmware updates
With consumer-grade devices, the owners are responsible for managing their devices, including data backup and for decisions around software upgrades. There is little to no control on software updates.
If data or the software used to collect the data changes mid-trial, this may render the consumer-grade device data non-usable for study purpose due to a lack of comparability between data before and after the software update, as this may change the way the data are processed on the device.
With medical-grade devices, the FDA recommends a Quality System (QS) regulation. The intent of the QS regulation is to ensure that there’s traceability within the software development lifecycle back to the design requirements.5 In addition to regulatory requirements, the medical-grade device manufacturers understand the requirements for controlling software/firmware version updates mid-trial to prevent or mitigate against data inconsistency related to changes made.
6. Transparency of data usage
Depending on licensing agreement, the consumer-grade device companies may share or sell the consumer-grade data to data brokers. The data brokers merge the device data with data from other sources to develop a consumer-grade profile that is used for business-oriented purposes.6 Thus, the clinical trial participant should be made aware of the likelihood of downstream usage of sensor data from consumer-grade devices; and approvals obtained vis-à-vis the participant informed consent form.
7. Availability of raw data and transparency of algorithms
The availability of raw data ensures that newer algorithms can be applied to older and existing data to derive newer outcomes. Several medical-grade device companies share the raw data collected via the device, and a few companies do so for an additional fee. The algorithms applied to the raw data can be available in the public domain and the sponsor may have the opportunity to choose the algorithm applicable to the study.
Raw data from consumer-grade devices is often not available and, whilst they can share the basic principles of the measurements, the algorithms used are proprietary. Additionally, should a consumer-grade company change the algorithm used mid-study, it could lead to lack of consistency in the data that are produced, and the conclusions that can be drawn from them.
8. Integrations with DHTs
As medical-grade device manufacturers are aware of drug development processes, they tend to have higher capability for integration into additional DHTs, such as electronic data capture (EDC) systems. Developers of other DHTs can collaborate directly with the medical-grade device provider; whereas for consumer-grade, they may have to collaborate with the third-party supplier of the devices. Integrations and interoperability between systems makes it easier for sites and participants to view and act upon the data generated from either sensors or other sources.
9. Operational and wrap-around services
Training for sites and participants, user manuals, and a help desk that can respond to questions relating to the use of the device are important operational considerations and required wrap-around services. Providers of medical-grade devices and platforms tend to have support capability designed with the clinical trial use case (e.g., supporting sites and potentially clinical trial participants) in mind.
Consumer-grade device manufacturers may obtain FDA clearance/approval as part of a marketing strategy to the consumer-grade market.
They may have FDA 510K or CE Mark but not plan to use their devices in clinical trials. One consequence of this deficiency is that a clinical trial may need a complex tiered help desk structure if the consumer-grade device company is not operationally or technically equipped to take enquiries directly from a clinical study team.
10. Economic feasibility
Medical-grade devices are generally priced higher due to factors such as completed regulatory approvals and clearances, the high cost of research and development, and the costs of clinical studies. Whilst consumer-grade devices lean toward affordability, the benefit of higher cost medical-grade devices may outweigh commercial-grade devices when the majority of the other nine considerations are satisfied. Return on investment is realized when participants are compliant, and all data are usable.
Sensors and connected devices used in a clinical trial must be fit-for-purpose and the selection is based on an interesting matrix of considerations. The sensor solution should meet the high regulatory standards required of sensor, data from sensor, and the sensor suppliers.
The technology of the sensor solution should comply with all region-specific requirements for patient privacy and data security. Patient centricity guidelines include the sensor physical design and affordability.
Clinical trials can be ongoing for years and suppliers should be able to support trials that are longer duration and global trials. To support longer duration trials, suppliers should have operational capabilities such as business continuity plans and a sound financial footing.
The application of these principles steers us toward and makes a case for medical-grade sensors and FDA cleared measures. In clinical trials, a medical-grade sensor solution inspires higher confidence in a successful implementation and regulatory submission.7
1. Hebenstreit C. How Technology Is Helping Increase Diversity In Clinical Trials. Forbes. June 18, 2020. Accessed November 9, 2023. https://www.forbes.com/sites/forbestechcouncil/2020/06/18/how-technology-is-helping-increase-diversity-in-clinical-trials/?sh=450a6d81bf89
2. Why isn’t Apple Watch’s blood oxygen feature FDA-approved? Government Technology. October 7, 2020. Accessed November 9, 2023. https://www.govtech.com/question-of-the-day/why-isnt-apple-watchs-blood-oxygen-feature-fda-approved.html
3. U.S. FDA. Deciding When to Submit a510(k) for a Software Change to an Existing Device. Guidance for Industry and Food and Drug Administration Staff. FDA.gov. Draft guidance. Published on October 25, 2017. Accessed November 9, 2023. www.fda.gov/media/99785/download.
4. U.S. FDA. Digital Health Technologies for Remote Data Acquisition in Clinical Investigations. FDA.gov. Draft guidance. Published December 2021. Accessed November 9, 2023. www.fda.gov/media/155022/download.
5. FDA Draft Guidance: Remote Data Acquisition in Clinical Investigations. Faegre Drinker. Web publication. Published March 10, 2022. Accessed November 9, 2023. www.faegredrinker.com/en/insights/publications/2022/3/fda-draft-guidance-remote-data-acquisition-in-clinical-investigations.
6. Izmailova, E.S., Wagner, J.A. and Perakslis, E.D. (2018), Wearable Devices in Clinical Trials: Hype and Hypothesis. Clin. Pharmacol. Ther., 104: 42-52. https://doi.org/10.1002/cpt.966. Accessed November 9, 2023.
7. Khurana L. Beckstrom K. What Are You Wearing? Applied Clinical Trials. Published January 2017. Accessed November 9, 2023. https://www.appliedclinicaltrialsonline.com/view/what-are-you-wearing