A physician-researcher details why new digital health products need to be informed by a combination of academic insight, industry innovation, and patient perspectives. Here's what he believes is lacking today and how the industry can move forward.
Michael Cohen-Wolkowiez, MD, PhD
Digital health technologies are enjoying widespread popularity, and the consumer market for these products has seen strong growth in recent years. Recent data from Pew Research shows that roughly 80% of U.S. mobile phone users own a smartphone, and a 2015 national survey of mobile phone users found that nearly 60% had downloaded a health-related app. The field of wearable biosensing technologies has also expanded in tandem with the growth in apps, with fitness trackers and other wearable technologies becoming increasingly ubiquitous in the last several years.
This rising tide of digital innovation has caught the attention of physicians and researchers like me, who are interested in harnessing these products for medical care and research. However, it’s important to note that while the field of consumer-oriented digital health technologies continues to enjoy widespread public interest and attract significant support from investors, not all of these technologies have been deliberately crafted to endure the rigors of the closely scrutinized, heavily regulated healthcare and clinical research environments.
Doctors, of course, are no strangers to using the latest technology to guide patient care. You could even say that we were early adopters of some of the first wearable health-tracking devices, although I’ll admit that a Holter monitor may lack the sleek appeal of current fitness trackers. However, the current wave of personal health technologies and applications—portable, affordable, networked, and capable of generating enormous streams of data around the clock—have the potential to liberate medical telemetry from the confines of clinical settings. In many ways, this represents the fulfillment of a dream long anticipated by medical technology visionaries: the capability to continuously capture high-quality data on states of health and wellness across enormous populations as these patients and healthcare consumers go about their daily routines in their accustomed environments.
Indeed, these broad aims are deeply intertwined with national initiatives now under way, such as the Precision Medicine Initiative and many others. Each of these programs relies to some degree on being able to tap into digital resources to craft better approaches to patient care, public health, and clinical research, simultaneously improving the quality and applicability of medical therapeutics while also bending the cost curve of medical research and development into sustainable territory. At the same time, however, these technologies present significant challenges that must be reckoned with if they are to truly benefit patients and the public.
The State of the Art, and What It Means for Medicine
There’s no doubt that some health-oriented devices and apps have enormous potential, or that many bright and imaginative people are working hard to take innovative concepts from the idea stage to actual reality. However, when we look at these technologies from the perspectives of physicians and researchers, often the “cool factor” of a new tech gadget or app is not sufficient for rapid implementation in the clinic, and more serious considerations quickly come into play. How trustworthy are these devices and apps? Are they accurate and reliable? Have their data and modes of operation been validated, and if so, by whom and in what patient groups or healthy populations? Are their data streams secure and interoperable with other existing systems and electronic health records? Can we examine how they work, or is the code or design hidden behind a proprietary “black box”? Are they safe for patients and consumers to use and do they have the appropriate privacy data protections? How do patients and consumers feel about these technologies, and the sharing and re-use of data gathered by them? And finally, is the industry working to meet the most critical healthcare needs or primarily chasing the consumer market?
As both a practicing physician and a researcher involved with early-phase therapeutic, device, and technology evaluation in an academic setting, I see several issues at play in the arena of consumer-oriented digital healthcare technology, some of which bode well in terms of improving care and facilitating research—as well as others that highlight opportunities for further engagement and education.
Developer Interest Abounds
There is clearly strong interest in creating applications for the healthcare sector among technology developers, and the growing procession of such technologies, whether marketed to the public or showcased at venues such as the Consumer Electronics Show (CES), reinforces this impression. However, there is also a substantial grey area regarding technologies that are being created strictly as consumer products versus those that have an actual healthcare or medical application. Where a piece of technology falls on this spectrum can have more to do with how it’s used than what it is. For example, if my wearable device is providing feedback that I use to shape my fitness routine at the gym, that’s one thing. But if that device is being touted as something that can guide medical or healthcare decisions, that’s a very different scenario, and any healthcare technology whose use could potentially pose a risk to the user’s health is yet another. I think I speak for most physicians when I say that we would want to understand and evaluate the data behind testing and validation, perhaps in multiple different healthy and patient subpopulations, before we would consider using a novel technology for medical applications. We would also be interested in data privacy protection measures and detailed information about the data flow, as well as potential interactions with established medical devices and electronic health records.
Development Activity Doesn't Always Track with Actual Medical Need
One aspect of consumer-based technologies—including the growing number of those focused on health, fitness, and wellness—is that they often seem to follow market trends in terms of their focus. For instance, when I attended a recent convention that included demonstrations of new consumer health technology, I noticed that many of the devices or applications were oriented around tracking various aspects of a person’s health — blood pressure, hydration, or mobility, just to name a few, and devices utilizing sensors that tracked vital signs in some form were well-represented.
Many of these technologies are of course potentially useful, not just for individual users who want to keep an eye on some aspect of their health, but also for healthcare providers, who could use the information from such devices to tailor care for individual patients or intervene in the event of an alarming signal (for example, a sudden and large increase or decrease in blood pressure). At a larger scale, we can also envision ways to use the flow of data from these devices for clinical research and for public health purposes. Indeed, the ability to leverage this “real-world data” is a topic of intense interest right now in the worlds of healthcare and medical research, as are new approaches to assistive technologies that harness recent improvements in imaging and audio technologies to help offset visual or hearing impairments.
However, the downside of developers flocking to a particular area can mean that resources and efforts are funneled into creating an array of “me-too” products. Although it’s good for patients and consumers to have options and competition can drive improvements and lower costs, the danger is that other, more challenging areas may be neglected in the pursuit of a market niche perceived as safer or more predictable.
For example: I think many clinicians would welcome an increased emphasis on rapid diagnostic technologies—particularly those that can be obtained non-invasively. An affordable and easily portable device that provided real-time, reliable measurements of glucose without necessitating a stick, or one that could accurately and precisely distinguish a viral infection from a bacterial infection, would have enormous appeal. Mature and trusted technologies of this kind would give patients and doctors powerful means for monitoring and controlling acute and chronic conditions. Although work in this arena appears (from my perspective) to be trailing existing needs, it may also be that companies developing such diagnostics have not typically participated in consumer-oriented information and marketing venues, and information on recent products and developments might be available elsewhere.
Domain Knowledge and Experience Is Lacking
When viewed from the perspective of clinician and researcher, the existing landscape of consumer-oriented healthcare tech shows a wide spectrum of knowledge and expertise. Some technology companies (typically, the more established and experienced ones) have a comprehensive understanding of the clinical environment, regulatory considerations, clinical trials, healthcare application, and payer acceptability of their product or device. Others, however, lack fundamental clinical trials knowledge and have little overall familiarity with the complex and rapidly evolving body of regulation governing the development and approval of products used for medical and healthcare purposes. Ideally, the skills and know-how needed to conceive and build a successful product prototype should be paired with those required for designing, conducting, analyzing, and publishing a clinical trial that provides the kinds of data needed to support clearance or approval for marketing from regulatory authorities. For example, companies frequently assume that the regulatory burden for app or device approval is high, when in reality (depending on the circumstances) it may not be. Having that approval can have important implications both for marketing and for the product’s acceptability to healthcare professionals. In addition, regulators are open to discussions about particular product development strategies and plans—an approach companies should consider if the pathway for approval is unclear.
Another key dimension relates to how physicians and researchers will be able to take these streams of data and place them in a context that allows them to be applied to clinical care and research. The end goal is to be able to leverage “digital biomarkers” comprising a constellation of digital data acquired directly from the individual through wearable biosensors. This can be used, sometimes in combination with other measures, to predict risk or guide treatment, ultimately yielding an “electronic phenotype” that correlates with particular states of health or disease. However, this is a complex, multi-stage process. Before any biomarker can be used for patient care or research applications, it must first be validated in one of several rigorous qualification processes that ensure the biomarker actually measures or correlates with the condition or outcome of interest. An additional challenge in this space is that with multiple products vying in the marketplace, there must be an assurance that these different products can measure the same phenomena in the same way—in other words, shared standards will be essential.
If technology developers want to make successful inroads in the healthcare space, they will either need to acquire highly specialized expertise in the clinical research and regulatory arenas, or else partner with groups or institutions capable of supplying that expertise. Another factor to consider is that as “patient-centeredness” becomes an increasingly important consideration in therapeutics, developers will also need ways to tap into the perspectives of patients and their advocates for guidance regarding the development and use of new technologies that serve patients’ needs and preferences. This may prove an especially salient issue as we drill down into the details of how consumers and patients interact with these technologies. For instance, although fitness trackers are now in wide use, researchers who examined willingness of patients within a health system to share data from personal fitness trackers found that only a small proportion followed up on an invitation to make such data available to their healthcare providers via a patient portal. This suggests that better characterization of user preferences regarding using such technologies and sharing access to data collected by them are needed. Indeed, the question of who has control or ownership of health-related data, and the uses to which those data are put, are critical issues for patients and consumers. Demographic and socioeconomic factors are also likely to result in differential uptake of these technologies. Finally, as systems for harnessing electronic health data mature, developers will need to pay keen attention to emerging standards for data curation, storage, transfer, and security.
The Way Forward
There are some genuinely exciting developments in the world of digital technology that hold great potential for both patients and their healthcare providers. Having recently had the opportunity to engage directly with some of the individuals working on these products, I was particularly encouraged to encounter both curiosity and openness to engaging with people from outside of the immediate confines of the tech sector. My hope is that as this arena of technology grows and matures, it will also be able to benefit from the kinds of knowledge and experience that clinicians, researchers, patients, and regulatory experts can bring to the table, while at the same time the clinical care and research communities stand to gain from the insights and expertise of seasoned technology developers. Such mutual exchanges of knowledge, experience, and viewpoints could help drive and accelerate innovation in both product development and in approaches to clinical trials, ultimately providing the kinds of high-quality data that can benefit patients, researchers, and the public.
Michael (Micky) Cohen-Wolkowiez, MD, PhD is director of the Early Phase Clinical Research Program and director of Pharmacometrics at the Duke Clinical Research Institute (DCRI).
[Image courtesy of GERALT/PIXABAY]