Health data at hand: How wearable medical devices are shaping patient care

Wearable devices already are amassing data that allow physicians and patients to gain new insights about human health.

Even with developments so far, in some ways, physicians, device makers and computer programmers are just getting started in the field of interfacing patients and the technologies that can tell them what’s happening in their bodies.

For health care policy, there is high level excitement about the potential for wearable technology. Health and Human Services (HHS) Secretary Robert F. Kennedy, Jr., made national headlines last year commenting on the prospects of wearables to improve Americans’ health. This month, HHS’ Advanced Research Projects Agency for Health is reviewing applicant summaries for its new Delphi program, a 4.5-year initiative to support innovation in biosensor technology. Proposals are due in May and initial prototypes within two years.

Matthew Smuck, MD, is chief of physical medicine and rehabilitation (PM&R) and professor and vice-chair of orthopaedic surgery at Stanford University. He’s also founder and director of the school’s Wearable Health Lab, dedicated to studying health care technology that patients can put on as part of their treatment regimens. He spoke with Medical Economics about the lab’s research and how wearables are a growing part of medicine.

This transcript has been edited for length and clarity.

Medical Economics: What is the Wearable Health Lab?

Matthew Smuck, MD: The wearable health lab is focused on using primarily wearables, but other digital health tools as well, in order to measure human diseases that affect mobility. We're looking primarily at orthopedic diseases and neurologic diseases, and then using that measurement to develop more precision medicine-based approaches to improve disease detection, treatment, and prevention.

Medical Economics: In presentation at the American Academy of Physical Medicine and Rhabilitation (AAPM&R) Annual Assembly, one of the goals was to describe how wearable technologies are set to reshape the future of PM&R musculoskeletal care. How will that happen?

Matthew Smuck, MD: It's an interesting time right now. When I started work in this area, the possibilities were fairly limited, and that was in part due to the infrastructure of medicine. At that time in 2008, when I first started my wearables work, electronic health records weren't universally used, and that's necessary for digital tools to become commonplace in the health system. Systems weren't in place to allow for the transfer of large streams of data from personal devices into the health system, because there were no HIPAA-compliant servers and clouds that could store and transfer that data, and there were no systems for getting it into the medical record, and so on and so forth. But we've seen all that change over the past decade in particular, and in the Bay Area, where I work, we've seen a huge surge in the number of startups around medical technologies. That's not because the tech world has developed a new interest in health and health care — the interest has always been there. It's that the tools that can now solve some of those problems are finally in place.

So, like I mentioned, we're in a very interesting time. Wearables are going to help change the way we do things, really, in two primary ways. One is that they will provide more accurate information about what's happening to people. That's important because, generally speaking, the way the medical system assesses the impact of a neurologic condition or musculoskeletal condition is through a conversation between the physician and the patient. That's useful, but it doesn't make for very good comparisons from one person to the next, and it doesn't allow for very accurate assessments of change over time. In disease states where those accurate assessments are available — such as cardiac disease research or cancer research, where there are objective biomarkers of disease that can be measured and tracked over time — those sciences develop in an iterative process and improve. Over a decade, you can see drastic improvements that occur because of the ability to iterate one small improvement upon another, and they're not reliant solely on large discoveries. But in disease states like musculoskeletal health, where things are measured through conversations or questionnaires, the science is imprecise and doesn't allow for that degree of iterative improvement over time.

The second thing that will change is that we'll have more continuous information on people, rather than just occasional point-of-service information. That means we can make more rapid and more accurate decisions about how to help people either prevent a disease or respond to it.

Medical Economics: When you talk about some of those changes in technology that have taken place, really just in the last 10 to 20 years, it sounds like we may be at the beginning of some new areas of study within medicine.

Matthew Smuck, MD: We absolutely are. We're seeing it happen right now, and there are some early examples of these things working. I was part of a group at Stanford that ran a conference around digital health, and I led a group of folks who are interested in wearables and came from different perspectives. Through group discussion, these experts from different areas of medicine reviewed two different successful systems of care that use wearables. Our group included people from hospital administration, insurance, medical research, and practicing clinicians. As we reviewed these two effective systems, we distilled out the common features that were integral to their success. These early use cases involved management of blood glucose for diabetes and management of hypertension, and both of those programs were more successful than the usual clinical care provided in the traditional manner. So they provide early examples of how these things can work.

It's hard to summarize in just a few sentences, but if I could point out what made those systems successful, it was that the wearable product was directly related to some feature of the disease that was important to measure and monitor. For blood glucose, it’s measuring blood glucose and for blood pressure, it’s measuring blood pressure. We can do that now at home using wearable devices in ways that weren't possible 20 years ago. And then integrating that data into a health system, one that allows for appropriate management through an app and with clinical support, is what makes those programs successful. You can extrapolate that thought to other technologies and other disease states, and that's what we're doing in the wearable health lab around musculoskeletal and neurologic health.

Medical Economics: Health and Human Servicers Secretary Robert F. Kennedy, Jr., has expressed support for widespread patient use of wearable devices in the movement to Make America Healthy Again. With his support, how has that been received or accepted by physicians and researchers already working with this technology?

Matthew Smuck, MD: I think people are excited that there's support for using this technology. It is going to be part of the future of medicine, and understanding that at high policy levels is important in helping it move forward.

Medical Economics: Since 2022, artificial intelligence has gotten a huge amount of attention for use in medicine and many other fields. How is AI going to be integrated into physicians' work and patient treatments involving wearables?

Matthew Smuck, MD: I would say that's yet to be determined. I envision it being useful, and perhaps it will be useful first in helping find parameters or features from wearables that were undetected by traditional statistical assessments. That's one of the ways AI has been useful in research: pointing in new directions that weren't considered based on traditional statistical analyses and models. That, by definition, is surprising, so we'll wait and see.

I think the other way it'll be useful, and perhaps more meaningful early on, is in providing individualized recommendations. For example, one of my research collaborators performed an elegant study where they took one of these mobility-limited populations — I talked about lumbar spinal stenosis — and provided them with recommendations about increasing the number of steps they take to become healthier and provided them with recommendations on places they could go to get more steps or things that they could do to get more steps withing their day and within the context of where they physically were geographically at any point in time. For example, a wearable program trying to promote physical activity might say, "Hey, I notice it's been several days since you went to the grocery store. If you walked to this store today instead of driving, you would gain this many steps." Or suggesting nearby parks, or different routes to work — things that offer real, simple solutions that are personally meaningful. AI is a tool that can make those things much easier to deliver, because when that study was performed, the underlying system had to be custom-built and took a lot of time. But AI can make features like that far more capable and easier to launch, produce, and maintain.