Camntech's MotionWatch

In February of this year, the FDA cleared two wristworn activity trackers from British company Camntech: non-smartphone-connected, but highly accurate devices intended for use in clinical trials.  In July, Australian company dorsaVi followed suit, obtaining clearance for its ViMove sensor system for detailed activity tracking. Meanwhile, Orthocare Innovations, a company that has long held FDA clearance for its StepWatch device, created a spin-off company called modus health to market their device, hitherto used mainly for research purposes, to clinicians. Even MC10's recent partnership with biopharma company UCB held up movement tracking in patients with neurological disorders as a possible focus of the collaboration.

For specialists whose practice revolves around patients' mobility, such as physical therapists, equipping patients with an activity monitor gives them a view into patients' daily lives they've never had access to before. But consumer trackers don't always cut it in these situations, either because they lack regulatory clearance, or because they're not sensitive enough for a particular use case. This has left a vacuum that companies like modus and dorsaVi are rushing to fill. 

Although their technology has existed, and even been in use, for years, the makers of clinical-grade activity monitors are suddenly stepping up their game, and as a result more and more physical therapists and neurological disorder specialists are beginning to use the technology in their clinics. This trend seems to have two major, converging drivers: a need for more quantitative data in an age of new payment models, and a trickle-down effect of the near ubiquity of direct-to-consumer activity trackers.

Why physical therapists like activity tracking

Dr. John Halamka, the CIO and a practicing emergency department physician at Beth Israel Deaconess Medical Center, told MobiHealthNews in a recent interview that his father's death helped him realize the value of continuously-gathered, patient-generated data.

"He had multiple sclerosis for 23 years," Halamka said of his father. "And it turned out — I didn’t know this — but he was using steroids on the day before I would come visit him, to mask his symptoms so we would have a more meaningful time when I visited. If he’d been using a Jawbone, Fitbit, or some other device, I would have seen that he walked 50 steps a day on average, every single day, except the days I visited, when he walked 500. ... That’s an example of where getting telemetry from the home would be extraordinarily important, not just for doctors, but for other carers or caregivers, family members. It was very hard to get an objective sense of his activity level just by talking to him or asking him how he feels."

dorsaVi's ViMove

That's exactly what most physical therapists do now: to assess a patient's mobility, they rely on self-reported data from patients or parents, combined with in-clinic assessments, such as video taping the patient walking for six minutes on a treadmill. Self-reported data is subject to both intentional and unintentional bias: patients who feel guilty about not moving enough might inflate their results artificially, or might just not have a good idea of how much they moved.

As for in-office clinical tests, that information is still valuable, and it's not likely to go away any time soon. At the same time, it doesn't always give the complete picture, according to Dr. George Fulk, a physical therapist, researcher, and professor at Clarkson University in Potsdam, New York.
Correction: A previous version of this article misstated Dr. Fulk's university affiliation.

"One of the nice aspects of having this information from activity trackers is you can really see how active they are outside of the clinic, and sometimes it’s a mismatch both ways," Fulk told MobiHealthNews. "People might perform really well on the tests you do in the clinic, and then you see their activity tracking data and they’re not really walking much at all, relatively speaking, compared to what you might have thought based on the clinical results. And then other times it's the other way around, where you have a patient that you might think based on their clinical result didn't perform that well, but they’re more active than you would have thought [when you look] at what they’re really doing."

Another physical therapist, Dr. Kristie Bjornson, compared take-home activity monitors to Holter monitors, which drastically changed how cardiologists assessed their patients a long time ago (and which are gradually being replaced by mobile health devices themselves). Bjornson is a pediatric physical therapist and an associate professor of pediatrics at the University of Washington.

"If you take a look at cardiac rehab, they’re getting more and more sophisticated and using that as the information to monitor their patients," Bjornson said. "So I’m expecting in 10 years or so we will do that as well. ... I, as a physical therapist, I would love to know how the kid’s doing walking at 9 in the morning, versus 5 in the afternoon, or Monday versus Friday after a full week of school. Is he functioning the same, is he more fatigued, and that’s the sort of thing an activity tracker can tell me. It’ll be very similar to how cardiologists use Holter monitors on their kids. "

modus health's StepWatch

For physical therapists whose specialty isn't walking, activity tracking might actually be even more valuable, because it allows researchers to draw a link between day to day movement and other physical ailment complaints, like lower back pain. That's the specialty of Rob Laird, who works at a clinic called SuperSpine in Australia.

"Up until recently there was little progress in the ability to accurately work out any reliable relationship of movement to lower back conditions," Laird told MobiHealthNews in an email. "But with the advent of increasingly smaller, wearable sensor technology such as the ViMove device from dorsaVi, clinicians have a much better ability to look at how real life movement may impact on musculoskeletal conditions. ... Lower back pain in particular is an enormous economic burden for society, and is such a common problem. If movement is related to lower back conditions, then smarter and more accurate ways of measuring movement is very important." 

When a Fitbit won't cut it

In the clinical setting, Fulk says there are two different use cases for activity monitors. One is as an assessment tool, either for sussing out the best intervention for a patient or evaluating how their current therapy is working out. The other is as an intervention itself: Just as consumer tracker devices are meant to motivate behavior change by tracking it, a patient whose therapy requires them to move a certain amount between visits might benefit from a device that gives them feedback on how much they're moving.

The latter is an area where consumer devices have an edge over many clinical devices -- because the clinical devices were originally designed for research, their data isn't always easy for the user to access. Notably, that's one thing modus is working on changing for future versions of the StepWatch, which currently has to be plugged into a PC to download the data.

"We did some research [with the Fitbit] and saw that it was relatively accurate with some people with stroke who could walk at faster speeds, and it’s easy," Fulk said. "It’s something they can put in their pocket, they can look at it and then bring it back to the clinic."

Fulk said that consumer activity trackers have become so popular lately, patients are even showing up with them in the clinic, asking if they'd be helpful in their treatment.

"I think in the past two years, [there's been an] explosion on the market of these more consumer, patient activity monitors like the Fitbit, Nike Fuelband, there’s one called the Shine, there’s just thousands of them and then apps you can download," he said. "And some of them are even free, the apps you can put on your phone, so I've seen those a lot in the past two years ... and sometimes patients even come in with a Fitbit before I’ve even seen them and say ‘How do I use this?’ or ‘I’ve been using this as a pedometer to track how I’m walking.’ Probably the average person is using it more than in the clinic, so that’s a good opportunity for a clinician to start using them more because our patients are already bringing them to us."

Halamka's upcoming patient-generated data project at Beth Israel will allow consumers to use their own over-the-counter activity trackers to give their clinicians data on their day-to-day movement patterns. Similarly, the Mayo Clinic has distributed Fitbits to cardiac rehab patients in a research study. Consumer wearables are cheap and user friendly, so physicians are happy to use them when it makes sense. But there are also limitations to devices designed for tracking the movement of well people.

"My understanding is a lot of the commercial devices, ... the algorithms that analyze the signal pattern, determine what’s a step and what are different activity metrics, those were developed for healthy people, so they may not necessarily work for people who move in different ways or walk slower," Fulk said.

Bjornson added that, especially for children, a good tracker for people with pathological walking patterns needs to not just be accurate, but be able to be calibrated as a child's walking pattern changes.

The Fitbit Flex

"A child might walk on his toes with his knees extended, and by the time he’s 10 he’s walking flat-footed and shuffling with a crouching pattern," she said. "That’s a very different kind of walking and the other devices on the market can’t be calibrated to pick up that walking. I think that’s the biggest take home for me for the StepWatch, why I’ve used it clinically in research. I can still monitor that child efficiently by being able to calibrate."

Many clinical activity trackers, like the ViMove, go further by tracking things that the over the counter devices simply don't.

"Clinical activity trackers can measure angle, forces and electrical activity of muscles that, for example, can be used to monitor change or may help clinicians to make more informed evidence based decisions," said Laird. "Each device has its place. When we want to encourage general activity a consumer tracker is helpful but if we need greater detail and want to personalize and measure the progress of these patients before and after we conduct an intervention then a clinical activity tracker is a really useful device."

Reimbursement and payment

The clinicians who are using activity trackers in the United States now mostly aren't getting reimbursed for them. Many of them started out purchasing the devices for research and started using them in the clinical setting subsequently. Bjornson says in her clinic the cost of wearable monitoring devices is rolled up in a general budget for assessment tools. Fulk said, similarly, that there's no real way to get reimbursed at present.

"There really isn’t any way to get reimbursed for something like that now," he said. "I use it more as a tool that the clinic already owns. We might have some piece of equipment we’re using to measure your strength, or a treadmill. We use it more as a tool that’s available in the clinic, so the actual device itself isn’t something that would be reimbursable."

ActiGraph activity monitor

But because the devices provide an objective measure of the patient's movement, they can be invaluable to a clinic under an outcomes-based payment plans, that requires physicians to be able to demonstrate results.

"I think for clinicians to start using it, one, it has to be pretty user friendly, and, two, they have to get into the idea of using it for documentation for reimbursement," Bjornson said. "I think down the road there’s going to be more and more demand for efficacy or effectiveness and I think that activity tracking, whether it’s walking or physical activity, is an excellent way of documenting the effectiveness of a physical therapy intervention in a child. Because the whole point is affecting their mobility in day to day life, but [right now] most of our outcome measures I report to the funding source are based on things I collect within my clinic setting."

Future adoption of activity tracking technology is also going to depend on devices that live up to all clinicians' needs, whether that comes through consumer devices becoming more clinic-friendly or clinical devices becoming more like the consumer ones. Laird, Bjornson, and Fulk all have wish lists for what they'd like to see the technology do.

"[W]earable technology will continue to evolve," Laird wrote. "Sensors will get even smaller; software will become more consumer friendly and hopefully research will identify more clearly how movement biofeedback might help in the management of the lower back. I would love to see iOS and Android apps connecting to the clinical activity tracker systems, and I’m sure this is likely in the future."

"It would be nice if, if it got lost, we could track it," Bjornson said. "Quite a few families lose them. I had one family that [found it] two years later and they gave it back. Other than being able to sync mobile devices to a computer or something, [I'd like to be able] to do an automatic download to my dataset without having to bother the family, but also having the ability for it to give feedback to the child and the family."

"Potentially having more GPS data to see where people are walking would be nice to have from a research perspective," Fulk said, "so we can see what’s the impact from a broader perspective that physical therapists do, to really make a difference with people that access their community more often."

Overall though, whatever form it takes, it seems that the activity tracker as a clinical tool, is here to stay.

"I think we will see an explosion of growth in this area," said Laird. "Technology will allow clinicians to 'see' much more clearly how people move both within the clinic as well as in real world situations such as work, sport and home activity."