Digital Health Lacks Data Dimensionality for Clinical Trials

In a previous interview on digital health, Dr. Michelle Longmire, CEO of Medable, discussed advancements in clinical trials. In this interview we want to add the perspective of a clinical investigator and digital health expert, Dr. David Lee Scher, founder of DLS Healthcare Consulting, and cardiologist at The Heart Group of Lancaster/Penn Med on how digital health is changing in both healthcare and clinical trial settings.

Moe Alsumidaie: How are wearables and digital devices morphing?

David Lee Scher: Digital health started out as a cottage industry and is now dominated by big corporations; for example, but initially, there were only small companies in the industry but now Apple is now leading the way in wearables. Additionally, I observe that digital health tech is becoming ubiquitous and the awareness factor has been surmounted, as evidenced by many of my older patients wearing fitness tracking devices; I don’t have to explain to patients what these technologies are. However, there are some failed outcomes in the evolution of digital health, which introduce negative perceptions that can detract researchers from using digital health in clinical trial settings. For instance, a big clinical study concluded that an increase in the number of steps via Fitbit measurements did not translate to weight loss. Nonetheless, the study did not evaluate other factors that may contribute towards weight loss, such as smoking and sedentarism.

MA: How are wearables being incorporated in healthcare settings?  How about clinical trial settings?

DLS: I am observing that there aren't a lot of new wearable devices, but, rather the expansion of the use of wearables in healthcare; from a professional standpoint, my practice has moved into wearables, as I am using water-repellant wearables to monitor patients’ heart rhythms for a week instead of having the patient wear a traditional heart monitoring device with patches connected to a monitor for 24-48 hours. I don’t believe there has been much penetration in clinical studies since wearables add a layer of complexity to the study. Additionally, the rudimentary data obtained by these wearables might not be relevant to most clinical trials; for example, vital signs like heart rate or rhythm might not always relate to drug studies. There are some Pharma companies like Pfizer and Sanofi which have embraced digital technologies into their clinical trial strategies.

MA: What functionality would make wearables more applicable towards clinical trials?

DLS: Devices that would collect more information would be very useful in the future, for example wearables that can sample electrolytes, degrees of hydration, and temperature through skin sensors. Wearables providing multiple data points simultaneously, and hence, broader dimensions of data collection. In my opinion, the lack of appreciation of the value proposition of multidimensional wearables is the reason these wearables have not hit the clinical healthcare and clinical trial industries and have remained in the consumer realm. Moreover, many wearables do not structure data in a way that is required for clinical study analysis (i.e., CDISC standards). Although there is some clarity and guidances offered by FDA regarding wearable validation (i.e., IMDRF), there is no specific guidance regarding digital health implementation and data collection in clinical trials. Another point is the lack of analytics tied to the deluge of provided data. In addition, more data is many times not desired by pharma because it leads to more scrutiny by the FDA.

MA: What will studies look like 10-15 years from today?

DLS: I would hope that patient-derived data would be an integral part of clinical trials, and hope that the analytics would be there to support study outcomes. The use of digital health would make clinical trials safer, quicker, with better quality data, and faster data collection. If you had adverse events that get to the data safety monitoring board, you could stop the trial much faster than when patients would come in for the visit or report adverse events through traditional methods. You could potentially reduce patient harm and enhance the well-being and rights of subjects. The utilization of blockchain technology, which has nothing to do with wearables but that could take the wearable data and make it traceable, verifiable and safer from a security and privacy standpoint, might find utility for sponsors, clinicians, and patients. This can also be shared and accessed in ways that it can't today and would be a phenomenal jump in the quality of clinical study data and processes. Another addition could be the use of digital biomarkers. Biomarkers don't necessarily translate to traditional vital signs or monitoring data points but are very personal and can also be used for clinical relevance. For example, NQ-Medical is a technology company which creates a neurological biomarker based on the way one interacts with a touch screen on a smart phone, tablet, or a computer keyboard. Technologies that leverage data, which can be collected and used in a totally unobtrusive and passive way with the individual, are going to be part of the future of clinical trials. They touch our daily activities without the need of either manual data entry or site visits. It goes without saying that CROs are going to have to find new revenue models because a lot of the work will be overtaken or substituted with such technologies, resulting in a marked decrease in the cost of clinical trials.

Moe Alsumidaie, MBA, MSF is Chief Data Scientist at Annex Clinical, and Editorial Advisory Board member for and regular contributor to Applied Clinical Trials.