Within the last decade, companies across pharma and the eClinical space have increasingly adopted wearable devices to collect objective, continuous, and real-time data from participants undergoing study interventions.
These devices streamline electronic clinical outcome assessment (eCOA) by reducing the need for study participants to self-report data and lowering the burden on these participants to visit study sites in person. Clinical trial sponsors can also realize an increase in longitudinal data quality, especially when minimizing bias from sporadic measurements.
However, before deploying wearable devices in a clinical trial, eClinical providers or pharma companies must validate these devices to ensure they function as expected and will meet the unique needs of each study to which they are applied.
But that requires expertise across domains like software development, eClinical data collection, and digital health. Below, we offer some insights into best practices for validating these devices to collect high-quality clinical trial data and accelerate development timelines.
#1 Choose devices for study-specific endpoints
When using wearables to gather clinical trial participant data, sponsors must tailor their software design to the specific endpoints they plan to evaluate. eCOA connected devices should not only qualify as fit-for-purpose within a clinical trial context, but they should also be amenable to the relevant scientific hypotheses being tested. These devices must be capable of gathering quality endpoint data in a scientifically or clinically usable format.
For instance, when studying sleep disorders, consumer-grade wearables like wrist-worn actigraphy devices may be cheaper to procure than medical-grade ones, but they may only be able to capture baseline trends like sleep duration or a user’s heart rate. As such, they lack the scientific rigor to collect large, complex data that would enable clinical trial investigators to comprehensively study the effects of an intervention on participants’ sleep behavior.
Even if these connected devices could capture comprehensive data from these participants, they cannot meet the precise scientific and regulatory demands of clinical research because their proprietary algorithms have not been clinically or analytically validated.
On the other hand, medical-grade devices typically use software with longer firmware lifecycles than consumer-grade ones, making them ideal for multi-year trials during which sponsors aim to collect longitudinal data from the same participants from start to end of the trial.
However, the solution isn’t simply to plug a medical-grade device into a clinical trial setting.
It’s important to choose a wearable device that’s validated for key factors like variations in the patient populations participating in a given study or the anticipated length of that trial. Short of doing so, sponsors may run into unforeseen costs, such as replacing devices midway a trial which could impact endpoint standardization.
#2 Assess regulatory compliance for connected devices
Although wearables must meet certain regulatory requirements before their deployment in clinical studies, these regulations are consistently evolving—making compliance challenging for pharma companies that lean on eCOA data to make clinical decisions about trials.
For instance, the definitions of medical devices vary between key frameworks like the European Union Medical Device Regulation 2017/745 (EU MDR) and the United States Food Drug and Cosmetic Act (U.S. FD&C Act), which means clinical trial sponsors operating these devices across the EU and the U.S. may have to optimize their compliance processes accordingly.
Whereas factors like 510(k) clearance are critical to validating the technical performance of wearables by establishing their functional equivalence to devices already legally marketed for the same use, eClinical providers and pharma companies should still test and validate these connected devices before using them to gather data within a specific trial setting. Focusing validation activities on outcomes like data integrity helps identify the gaps that may compromise eCOA data quality in the long run.
Beyond regulatory compliance, adopting security guidelines from adjacent frameworks like the Health Insurance Portability and Accountability Act of 1996 (HIPAA) helps secure sensitive clinical trial endpoint data as it’s wirelessly transmitted from wearable devices to the cloud (or other storage locations) for stakeholder access.
These wireless transmissions are susceptible to cybersecurity risks, which if undetected early enough, can compromise entire networks containing large amounts of sensitive data. Such validation is even more critical when partnering with third parties like clinical research organizations (CROs) to process data for clinical trials.
#3 Develop eCOA solutions with the patient in mind
Regardless of how long a patient participates in a clinical trial, it’s critical to design eCOA software with patients’ needs in mind. Wearables that seamlessly integrate into a patient’s daily routine or passively collect data can lower the burden of participating in a trial, especially over several years of continuous remote monitoring.
For example, an eClinical provider partner successfully implemented meticulous, patient-centered design elements into an at-home connected device used to study a specific type of migraine. Their team identified potential areas of patient disengagement and addressed them before the study, ensuring patients understood trial requirements and how to use the device properly.
Conversely, difficult-to-use devices can impact patient compliance, especially if the devices have technical issues that result in glitchy or buggy interfaces. Instead, clinical trial sponsors should design software that engages patients with usable, inclusive, and validated interfaces that minimize patient fatigue and lower the risk of disruptions to data collection.
While the patient experience is central to the design of eCOA solutions, keeping sites in mind ensures that the technology works well within the broader ecosystem of a clinical trial. Sites are crucial for operational success, data integrity, and providing ongoing support to patients, so developing solutions that are user-friendly for both groups will lead to better outcomes for everyone involved in the trial.
#4 Leverage proven third-party validation expertise
Faced with multiple development costs when deploying connected devices in their clinical studies, some sponsors may choose to handle processes like device validation internally.
However, organizations that develop their connected device technology for eCOA or other digital health applications internally are more likely to incur greater costs than those that partner with clinical and analytical validation experts. The performance requirements for these devices may vary from one study to another, meaning a sensor or monitor that worked in a similar study might not meet the needs of a closely related one.
For instance, a clinical study tracking atrial fibrillation in high-risk patients requires a heart rate sensor with a higher sensitivity and accuracy than one evaluating longitudinal resting heart rate in healthy young athletes. These considerations will help guide device validation and ensure sponsors choose the right devices to include in their studies.
The Future of Wearable Devices
McKinsey recently called out the potential of wearable devices as part of the trend in increasing secondary endpoints per trial for companies looking to maximize the values of their studies:
“…technologies enabling continuous patient monitoring—such as wearable devices—will support digital biomarkers as secondary endpoints. These will enable monitoring of nuanced, continuous real-time changes in disease progression and treatment response that traditional endpoints might miss.”
While promising, there are challenges to deploying wearable devices effectively. Successful eCOA clinical studies require well-designed, reliable technology (and connected devices) that work effectively to enable continuous patient engagement and reduce patient burden while minimizing impacts to data quality.
