Digital Measures of Health: What’s in it for Patients?

Digital measures offer new insights into human health but require complex, multi-stakeholder development. Are they worth the effort? Can they deliver meaningful health impacts?

At the SCOPE Summit in February, the authors led a panel in which they discussed digital measures and patient experience. This article summarizes the discussion and outlines the different stages in the development of a digital measure, detailing specific activities along the journey toward the finish line and ultimately, to meaningful impact for patients.

Caspar, 2, suffers from atopic dermatitis

Meet Caspar

Here, we introduce you to a two-year-old boy named Caspar. Caspar suffers from atopic dermatitis, and he needs to be covered in creams to stop his skin from itching. He doesn’t sleep well at night because of the itching which, in turn, disrupts the sleep of his parents and sibling. Caspar and his family are waiting for a therapy that will stop his itching and related sleep disruptions. How might a digital measure help a patient like Caspar?

Patients and measure definition

Patients often experience disease impacts beyond primary symptoms. Focusing only on Caspar’s skin issues—itching or redness—one might overlook the real burden: sleep disturbances. To avoid such blind spots, we must first listen to patients and caregivers. Are they sleeping at night? What are they not able to do because of their disease? How does this affect their quality of life? Conversations with physicians can also reveal important insights, for example, that atopic dermatitis patients are not often aware that they scratch themselves.

With this understanding, the next step is to define the concept of interest that could be measured and why this is meaningful to patients. While Caspar’s itching triggers sleep disruption, asking him to report itch levels isn’t feasible or reliable given his age. Instead, measuring scratching activity may provide objective insight to his itch and the scratch, but also insight into the sleep disturbances that he is encountering.

After defining what we can measure and why, we can then identify the tool that would work best in a naturalistic setting for that patient. Moreover, to understand a person’s subjective feeling about something, in this case the itching, we can use patient-reported outcome measures. But as stated earlier, this can be challenging in some populations. To directly measure an activity or function, we can use objective digital measures that capture the motion of the symptom. In Caspar’s case, measuring the scratching activity is the most meaningful concept to be measured; it directly correlates with sleep disturbances that impact him and his family, as often scratching can precede the sleep disturbance.

Tech validation: Are we doing it right?

Now that we know what to measure and why, let’s focus on how. We need to consider not just which technology to use, but also the right modality for the technology and the patient population while balancing analytical performance and patient usability. For example, while infrared cameras in a patient’s bedroom might provide great data, they are impractical because patients are unlikely to accept them. Wearable solutions such as wrist-worn devices or smart rings might be better solutions, but how do we select the right one?

Understanding patient preferences early minimizes issues in later stages of trials; we wouldn’t want to discover during a large Phase III trial that patients refuse to use a selected device. Coming back to Caspar, he is an active toddler and may not sleep well with a heavy device. A ring could be an option, but will it stay on at night?

Once the modality and solution are chosen, we need to generate evidence about its analytical performance in the target population. In Caspar’s case, a study could investigate how well we can measure scratching activity based on multiple devices worn by a pediatric population. Analytical validation ideally measures against the “ground truth,” which for Caspar, could be video recordings.

Clinical validation helps with understanding the new measure; for example, how it correlates with other known aspects of the disease. It helps in making decisions, such as classifying patients by scratching activity or understanding clinically meaningful improvements.

Without clinical validation evidence, we cannot make decisions or claims based on the data, which would hinder the ability to use the evidence for drug approvals. This work ensures we have the right solution and the evidence to demonstrate validity to regulatory agencies for acceptance.

Collaboration: How is it relevant to the patient?

Drug development is a competitive industry, and this already helps Caspar that there are multiple horses in the race; competition is a good thing. This context is key to understanding where the collaboration opportunities are and where they are not.

While companies compete to deliver the best interventions, there are opportunities for collaboration in developing digital measures. Pharma organizations can work together on defining, standardizing, and validating these measures; these activities are typically pre-competitive.

Efficient engagement with patients and regulators is crucial. It is impractical for everyone to talk to the same patients and physicians about their preferences, and repetitive conversations consume time from regulators. Consolidating efforts around measure definition and regulatory standards can reduce duplicative work. Validation work can be pre-competitive, especially if done outside drug development programs.

Efficient and ethical use of resources among pharma companies, academia, technology companies, regulatory authorities, and other stakeholders is essential but challenging. Still, there are great examples of multi-stakeholder collaboration. Digital Evidence Ecosystem and Protocols (DEEP), Innovative Medicines/Health Initiative (IMI/IHI), and Digital Medicines Society (DiMe) have been facilitating great collaborative work in recent years. We all have a responsibility to drive broader collaboration to help deliver better treatments to Caspar as quickly as possible.

How do regulatory policy and innovation impact the patient?

Pharma companies are working on interventions to help Caspar, but they need to prove that the therapy reduces scratching to get approval. Having the right therapies and technology is only part of it; regulators must be on board with the digital measure that supports the approval of new therapies. The field of digital measures is evolving, and while some guidance is available, it often leaves substantial room for interpretation. Many practical issues still need to be addressed on a case-by-case basis.

Harmonizing measure definitions across drug programs is challenging, but regulators are open to collaboration. Cross-industry approaches can help regulators by reducing duplicative questions from different companies. Agencies can collaborate to establish regulatory standards and expectations to streamline acceptance of digital measures.

There are great opportunities to develop standards and improve coordination. With the right standards, frameworks, and tools, we can unlock new insights into human health and understand disease and intervention impacts.

What’s next for Caspar?

While Caspar and his family wait for a new therapy, they are cheering on those of us working on new therapies, digital measures, tools, collaboration, and regulatory frameworks. The more effectively we demonstrate a new therapy’s ability to address Caspar’s nighttime scratching, the sooner he will find relief.

In summary, here is what we can do to help us move more quickly toward helping Caspar:

• Seek to collaborate pre-competitively.
• Be clear on what to measure and why, with input from patients and physicians.
• For the tech solution, find a balance with what works and what patients can use.
• Speak to regulators early on to get input on a proposed approach.

Caspar is already waiting for us at the finish line, so let’s think about the bigger picture, leverage the synergies within the ecosystem, and get the best therapies to him.

Kai Langel, Founder and CEO, DEEP (Digital Endpoints Ecosystem & Protocols) Measures; Martha Azer, PharmD, Associate Director, Regulatory Policy North America, Johnson & Johnson; Bola Grace, PhD, Professor of Practice, University College London; Scottie Kern, Executive Director, eCOA Consortium at Critical Path Institute; and Carrie Northcott, PhD, Head of Digital Sciences, Biomeasures Endpoints, and Study Technologies (BEST), Pfizer