Data Protocol and Wearables: How Trustworthy Is the Data?
Tracey Romero • Mon, June 19th, 2017
The importance of big data to medical and scientific communities is not a new concept, but the growing popularity of wearable sensors and their ability to collect data useful to doctors and researchers is presenting a mixed bag of opportunities and challenges.
According to the Smart Wearables in Healthcare, 2016-2030 report from research firm Research & Markets, the overall market for smart wearables within healthcare is expected to grow at an annualized rate of 13.6% over the next 15 years, but while wearables provide a lot of physiological and mechanical data, there are still concerns over the reliability and validity of the data.
At the recent American College of Sports Medicine (ACSM) annual meeting in Denver, Colorado, a panel on “What’s Next with Wearables? Industry Experts Discuss the Rapidly Evolving Technology Landscape and Its Implications for Physical Activity and Sport,” which included representatives from Mio Global, Medtronic/Zephyr Performance Systems, Adidas and Technogym, discussed some of the challenges and opportunities posed by wearable sensors for clinical care, research and sports medicine.
Stacey Burr, general manager, digital sports for Adidas discussed Adidas’ wearable products including clothing, wrist devices, and shoes, as well as their new fitness app called All Day which was developed with several partners, including the American College of Sports Medicine and Verily, Alphabet’s healthcare and life sciences subsidiary. The app collects data on movement, nutrition, mindset and rest, and based on that data offers tips which Adidas calls “Discoveries” on healthy living like workout routines or how to sleep better. The app is used in conjunction with the Adidas Chameleon, a lifestyle tracker.
Burr said, “We have a strong sports science foundation engrained in our DNA and we would like to partner with doctors and researchers on science-based technology to improve activity, sleep, nutrition and stress levels.”
Another one of the panelists was Todd Monrad, sales director, health, corporate and public for Technogym, which focuses on fitness equipment like their home treadmill My Run that can collect important health metrics from the user and make the data available on the machine and on any personal device (mobile, tablet, smart TV, PC) through their Mywellness cloud computing platform.
Monrad said, ““We embrace wellness on the go. We developed a software platform that allows the managing and monitoring of clients to create a remote coaching environment. It allows the ability to connect exercise prescription with actual physical activity.”
He added, “Our mission is to help people live a better life, innovate to engage end users and help them through their wellness mission. We would love to partner to monitor exercise prescription to determine if a patient is following the prescription We are also working on validating the equipment to use in a clinical capacity for patients instead of athletes.”
Mark Gorelick, chief science officer for Mio Global was also on hand to discuss to Personal Activity Intelligence (PAI) which uses an algorithm to make heart rate accessible and understandable so the user knows how much activity he or she needs to live a longer, healthier life.
Gorelick said that PAI was created from the HUNT database conducted at The Norwegian University of Science and Technology study with over 25 years of data and over 45,000 participants.
He added, “Studies show that keeping your PAI score above 100 can reduce the risk of cardiovascular disease and extend longevity.”
PAI is currently featured in the Mio SLICE, a heart rate activity tracker. Gorelick said that another benefit of PAI is that it is universal unlike step tracking and includes any activity that elevates heart rate. You just need to keep your PAI above 100 over a 7-day rolling window to know you are staying healthy which means if you prefer, you can work harder on some days and less on others.
Mark Santini, senior. global product manager, Zephry Performance Systems, Medtronic, during the panel talked about the Zephyr Performance Systems which includes a sensor, garments, communication and software that allows performance data to be collected, analyzed and displayed through Omni Sense software.
He said that the Zephry Performance System measures 6 key inputs that report on more than 20 biometrics which can help in the training of athletes, military and first responders and in emergency situations. It can also help researchers studying new vaccines, antibiotics and medication understand the effects on the human body so they can improve effectiveness and reduce unwanted side effects.
In a later interview with OTW, Santini added that for doctors and researchers “the raw data is the best place to start, but there is also an aspect of the software that is really helpful with exercise loading prescription especially before and after an injury. For example, how to avoid an ACL [anterior cruciate ligament] tear and make sure an athlete is not over trained. The data could be helpful with developing rehab protocol.”
The Science Behind Wearable Sensors
While there is still much work to be done on validating wearables to be used in healthcare settings, some research has already been conducted. In a study published in the March 2012 issue of the Journal of Sports Science & Medicine, researchers found that the use of Technogym’s MyWellness Key (MWK) accelerometer during a treadmill-based protocol can be helpful in identifying exercise intensities that are related to walking and running. And another study published in January 2013 issue of Clinical Epidemiology found that the MyWellness Key accelerometer had good validity during light and moderate walking, but there was some overestimation when it came to running.
The Zephry Performance System was featured in a study published in the January 2014 issue of PLoS One. The researchers used heart rate variability data from Zephyr monitoring tools to study the effect of prolonged stay at high altitude for autonomic functions.
PAI is also heavily backed by science. Ulrik Wisloff, Ph.D., a professor and head of Cardiac Exercise Research Group and K.G. Jebsen Centre for Exercise in Medicine at the Norwegian University of Science & Technology invented PAI using the HUNT database.
In the March 2017 issue of the American Journal of Medicine, Wisloff along with Carl J. Lavie, Jr., M.D., FACC, FACP, FCCP, FESPM, medical director of cardiac rehabilitation and preventive cardiology and director of exercise laboratories at the John Ochsner Heart and Vascular Institute in New Orleans and other colleagues investigated PAI’s impact on the prevention of cardiovascular disease and the promotion of physical activity and found that a PAI of 100 or greater was associated with significantly lower risk for cardiovascular disease mortality even in people with known cardiovascular disease risk factors.
In a new study that will be published in the July/August issue of Progress in Cardiovascular Diseases, Wisloff and colleagues also looked at PAI’s impact on sedentary behavior and cardiovascular risk factor clustering.
In an interview with OTW, Lavie said, “It is really a potentially very effective way for clinicians to know if people are getting enough exercise. And theoretically, if you tell your health plan you have 100 PAI, they could reduce your premiums because they know you are going to be protected against health consequences. There is potential use on a big scale for companies, health insurance plans, but also individuals.”
He added that “PAI has also been studied against the federal guidelines which is just the recommended time of moderate or vigorous exercise. And those who met their time but didn’t reach their PAI had a 27% higher cardiovascular mortality.”
“Now will this become a part of the federal guidelines? It takes a long time to create change, but an orthopedic or sports medicine doctor could use this to know if his or her patient is getting the right amount of effective exercise.”
Future of Wearables in Healthcare
One of the biggest things that came out of the ACSM Wearables Panel was a wish for more collaboration between the research community and companies manufacturing wearables. The audience stressed the need to address concerns over the reliability and validity of the data as well as data security and privacy. For instance, who owns this data which often gets sold to third-party agencies?
One audience member said, “Thousands of people uploading data makes researchers salivate but we can’t publish a paper without informed consent from participants. How do you get that?”
Other areas of concern were how wireless charging can affect the continuity of the stream of data and the need to find better ways to decode the data from wearables to give more personalized feedback that could be more meaningful to the user and his or her doctor.
Many of these concerns are shared throughout the research community. In an article in the February 2016 issue of PLoS Med, the authors wrote, “Use of consumer wearables in clinical populations is still limited. Currently wearables exist within a ‘grey area’ regarding user safety,” meaning that while there isn’t a lot evidence of potential harm in current literature, people could be given a false sense of security and misdiagnose themselves.
The authors also recommended the creation of a simple regulatory framework that “doesn’t suppress innovation but helps wearable devices become validated in the context of their health-oriented value.”