Without question digital health technologies are revolutionizing clinical development and execution. From electronic patient reported outcome (ePRO) systems to wearable devices, they can, when used properly, provide many benefits, including:
- Accelerated trial timelines
- Increased data accuracy
- A patient-centric approach to data collection
- Improved treatment adherence
- Increased ability to study rare disease
However, it is unwise to assume that a digital health solution is the right solution in every trial or for every aspect of a trial. There are many factors that must be considered before deciding how a study should be administered and what tools should be used.
Who is the patient population?
Rather than first asking “Can a digital solution work in this trial?” the initial question should instead be “Can a digital solution work for these patients?” Technology that makes the job of the sponsor or CRO easier while making participation more difficult for patients should, of course, not be used.
One of the more obvious delineations here is studies involving younger versus more mature populations. The former is likely comfortable with technology, having been born in an era when it is pervasive. The latter may find the use of digital tools challenging and even intimidating. Something as commonplace as a touch-screen interface, for example, may be unfamiliar to more mature populations.
And, of course, age is just one example of a factor that might affect the type of technology that can be effectively employed. There are many others. Patients with fine motor skill challenges or limited vision would find a small digital device difficult or impossible to use, for example. Ultimately, determining the appropriateness of digital solutions for a patient population is of paramount importance.
Can and should devices be integrated with the EDC?
In many cases, mobile health (mHealth) devices can be integrated with EDC systems. This can happen in one of two modalities. The first is full integration. Often used when immediate feedback is advantageous, it allows data captured by the device to be transmitted to the EDC in real time or when the device syncs. The second, appropriate when a large amount of data is captured and is not needed immediately, is batch uploading.
However, the fact that integration may be possible doesn’t necessarily mean it should be pursued. That decision should be based on protocol and technical considerations. Questions related to the protocol include:
- Is the data needed for eligibility assessments?
- Must the data collected from mHealth devices be checked against other clinical endpoints being captured in the EDC?
- Is it important that data be available in real time?
Technical considerations include:
- Does the EDC have an open application programming interface (API), which is required for integration?
- Does a member of the study team have the technical skill available for creating this integration?
- Can the EDC system accommodate the needed page type?
- Is there enough time to complete the integration?
In short, integration is most appropriate when mHealth data 1) is needed to make real time decisions, 2) is key for primary endpoint evaluation, 3) is needed to provide context for the assessment of other data in the EDC, and 4) is discrete and manageable.
Integration is not ideal when data 1) is not required for decision making during the study, 2) is not necessary for primary endpoint evaluation, 3) is being cleaned by the mHealth vendor outside of the EDC, or 4) is continuous and therefore voluminous. And, of course, integration should also be evaluated in the context of the study timeline and cost.
Is more data better?
Digital solutions have the ability to collect literally millions of data points per subject per day. This presents obvious opportunities but also a number of challenges.
For example, how great is the potential for true data signals to be lost in the “noise” of a huge volume of data? And, can the EDC system manage the volume of data that the mHealth devices will be delivering? Some of the more robust systems are designed to handle extremely large datasets, but in an effort to control costs, many studies rely on a more basic version that will be unable to accommodate all the data produced.
In some situations, it may be that less truly is more.
Are there creative solutions to existing challenges?
While the use of digital health technologies comes with very real challenges, in many cases, there are solutions for them. For example, EDCs can be designed to capture data only at certain intervals, making huge volumes of data smaller and more manageable. Or, the handling of data can be entirely outsourced to the mHealth provider, who in some cases will have data management and or biostatistics expertise in-house.
Digital health technologies are here to stay. With proper forethought on when, where, and how they should be used, sponsors, CROs, and patients can maximize the benefits they deliver.