7
Reflections and Key Takeaways
The workshop concluded with reflections and consideration of key next steps for the use of digital health technologies (DHTs) in drug development. Jennifer Goldsack and Joseph Menetski discussed the risks associated with the use of new modalities for collecting digital health data, the importance of patient centricity and education, the need for standards and frameworks for evaluation, and other barriers and opportunities.
CHALLENGES ASSOCIATED WITH THE USE OF DIGITAL HEALTH TECHNOLOGIES
With the new methods enabled by DHTs come novel data, but also concomitant new risks, Goldsack said. She emphasized the importance of finding ways to mitigate those risks while also taking advantage of these new data to their fullest potential. As new DHTs become more powerful and pervasive, it is important to consider whether the data they generate are being appropriately analyzed and correctly interpreted. Care should be taken to ensure that these new technologies are being deployed in an ethical manner. The need to apply an ethical lens to this work was introduced by Camille Nebeker and echoed by participants throughout the workshop. Strategies are also needed to deploy existing technologies without reinventing the wheel, given the large number of tools and the volume of digital data already available. The coronavirus disease 2019 (COVID-19) pandemic has underlined the need to deploy new digital technologies in practical ways. Goldsack described the COVID-19 pandemic as a pressure
test for the current state of digital measures. While there are a variety of capabilities available for DHTs, she said, these may not be ready for prime time. The workshop participants helped to shed light on some of the reasons why DHTs have not been instantly deployed and immediately ramped up in the same way that telehealth has, for example.
DIGITAL HEALTH TECHNOLOGIES TO ENABLE PATIENT-CENTERED DRUG DEVELOPMENT
Patient centricity was a consistent theme running through the workshop, Goldsack said. Multiple workshop participants emphasized that patients are at the heart of the development and deployment of digital measures. For these measures and tools to be equitably deployed, they should be developed in ways that ensure they perform equally well and are universally available across the entire population. Education is also foundational in improving equity and accessibility, she said. All stakeholders—from study participants to clinical trialists to institutional review boards—need to clearly understand the benefits, risks, and opportunities associated with these DHTs. She reiterated a call to action that was issued by Amy Abernethy and then echoed through the rest of the workshop’s discussions: the need to bridge the divide between data generated through clinical care and clinical research. Digital measures of real-life phenomena warrant commensurately practical strategies for collecting that evidence in the real world, in ways that are perhaps less controlled than traditional methods. There is often a forced conflict between privacy of participants and speed of data collection, Goldsack said, echoing remarks earlier in the day from Nebeker. Although the escalating pace of innovation in DHTs poses certain barriers to maintaining ethical standards in this space, it also heightens the need to do so from the outset of any study design or product development efforts. Participants dispelled myths around regulatory barriers being a limiting factor to rapid progress in the field—in fact, existing regulatory frameworks can provide solid foundations to build on.
Opportunities, challenges, and solutions for better engagement with patients were also discussed throughout the workshop. DHTs provide opportunities to build trusted relationships with patients and illuminate the patients’ experiences. However, the adoption of those technologies is hindered by the lack of capabilities to easily share data onward with clinicians and researchers and then return value to patients in a meaningful way, Menetski said. Potential solutions to these engagement challenges, which were highlighted by Yvonne Yu-Feng Chan and Camille Nebeker, may include (1) building trust and educating patients through health provider advocates, (2) using digital technologies to foster a com-
munity for participants to connect with each other, (3) using digital tools judiciously in specific contexts that are supported by human contact, and (4) educating the general public to build trust in the outcomes of DHT use.
STANDARDS AND EVALUATION FRAMEWORKS
Goldsack outlined several gaps that emerged throughout the workshop related to standards for evaluating DHTs. The field is currently lacking the standards and evaluation frameworks needed to perform the necessary verification, analytical validation, and clinical validation of DHTs (see Box 7-1). The dearth of well-established, gold standard reference measures is also a challenge for verification and validation. This is a particular concern in cases where no reference measure exists at all, as no standard method is consistently applied in those situations. More broadly, methodological approaches have not yet been standardized across the field. This undercuts the effectiveness of comparative analyses of the tools and technologies themselves, as well as the quality of the data that are captured and the interpretations that are derived from those data. Similarly, better standards are needed to define, describe, and understand the data collected by these technologies. In addition to fostering common understanding among stakeholders across the field, better data standards would enable more consistent and accurate reporting, evaluation, and sharing of the data. Goldsack noted that another critical missing component is a single “source of truth” that would facilitate linkages across datasets and allow actors in this space to infer meaning from each other’s work.
Although DHTs provide opportunities to expand and standardize the capture of data in clinical research, Menetski said that strategies will be needed to work within the current existing limitations of data quality
while also developing methods to improve the data quality (e.g., instrumentation improvements, supplemental data points, normative studies) and interoperability. Reflecting on comments heard earlier in the day, he suggested that the quality, validity, and reliability of data would benefit from community-wide standards for data capture and interoperability, a repository for digital health data, greater population diversity in data collected, and strong frameworks for data governance, as well as common methodologies for validating DHTs and for comparing digital metrics to clinical ratings.
Goldsack reiterated that these efforts to propel and expedite progress in digital technologies should not be reinventing the wheel; instead, they should learn from and build on the work that has already been done. For instance, the pharmaceutical industry has instituted several initiatives for driving standards and using devices and data in noncompetitive ways, such as the Metrics Champion Consortium1 and TransCelerate.2 She suggested several other initiatives that might be useful resources, including the Digital Medicine Society,3 the Clinical Trials Transformation Initiative, and the Open Wearables Initiative.
“PAIN POINTS” ACROSS THE DRUG DEVELOPMENT PROCESS AND SOLUTIONS DIGITAL HEALTH TECHNOLOGIES CAN PROVIDE
Menetski reflected on the differing “pain points” across the drug development lifecycle that speakers presented, and he summarized the opportunities that DHTs provide as well as the unique challenges that arise when they are used in clinical research. In discussions about disease characterization in the early stages of development, the lack of interoperability was a commonly cited barrier (see Chapter 3). The challenge of retaining study participants emerged as a substantial barrier in the contexts of recruitment, safety trials, and post-market surveillance. As put by Yvonne Yu-Feng Chan, “the Achilles’ heel of digital health technologies is retention.” Speakers in session 2 (see Chapter 4) highlighted the value of understanding what matters to patients and participants and reporting it back to them in order to help increase adherence and improve consistency. Challenges related to the lack of benchmarks and gold standard reference measurements for demonstrating that a new measurement is actually
___________________
1 Information about the Metrics Champion Consortium is available at https://metricschampion.org/mcc-risk-quality-scoring-tools (accessed May 17, 2020).
2 Information about TransCelerate can be found at https://transceleratebiopharmainc.com (accessed May 17, 2020).
3 Information about the Digital Medicine Society is available at https://www.dimesociety.org (accessed May 17, 2020).
measuring what it is intending to measure were discussed in the context of pivotal trials (see Chapter 5). Speakers discussed a concrete solution to this barrier; rather than waiting passively for a single entity to take the initiative, the entire community should take collaborative action to develop gold standards. Speakers also emphasized the importance that these efforts to convene around community-wide standards be inclusive of all stakeholders, from product developers to regulators to the teams of health care providers on the ground who are caring for people’s health and to patients and their representatives. At the same time, the bottleneck around validation standards could be addressed through a collective decision to accept the best available standards—and work to refine them—rather than allowing “perfect to be the enemy of the good” and continuing to operate without any standards at all.
Goldsack encouraged participants to be involved in the process of implementing the learnings from the workshop and help build an action-oriented body of work to advance the field. We are the community, she said, and we need to be the ones who develop the standards as soon as possible.
This page intentionally left blank.