Feature

Wearable devices are hot and getting hotter. The global market for wearables topped $2 billion in 2015 and was expected to surpass $4 billion in 2017, according to the market research firm Statista Inc.

Wristbands, watches, rings, clothing—you name it, there's now a digital version that can track how many steps you walked today, measure your heart rate or evaluate how well you slept last night. But now mobile technology is pushing new boundaries, aiming to measure emotional well-being as well as physical health.

And because we carry our smartphones everywhere, mobile phone applications have become de-facto wearables, too. A host of sensor-driven smartphone apps already claim to passively monitor stress or other facets of mental well-being.

"Mental health is the next frontier for mobile health," says Jiten Chhabra, MD, a human-computer interaction researcher at Georgia Tech who studies mobile health technologies and works to measure the effectiveness of these products. But when it comes to using technology to track mental health, Chhabra and other experts say, there are good reasons for consumers to tread carefully.

Digging for evidence

Tech companies claim that biosensor technology in the devices can detect stress and anxiety by tracking the body's physical reactions. But the hard evidence behind these claims is often sorely lacking.

"Go to the health and wellness category in a mobile store, you'll see thousands of apps, but the majority provide no evidence of the health claims they present," Chhabra says. "These frivolous health claims cast a shadow on the entire industry."

Some companies are tackling these credibility issues head-on by commissioning psychologists to study the effectiveness of their technologies. One such company is Spire, whose device clips to your clothing to measure your breathing patterns and detect periods of calm, focus and stress. When respiratory patterns indicate you're stressed, the system's smartphone application guides you through calming breathing exercises.

Spire, which sells for around $130, was born from a project that began at Stanford University. The company recently partnered with Stanford psychology researcher Alia Crum, PhD, and colleagues to study the device among employees of LinkedIn. Compared with a control group, employees who wore the device and followed the breathing exercises had lower symptoms of stress and negative affect. Over a 30-day period, people in the Spire group reported 27 percent fewer "anxious" days and 35 percent more "energetic" days, according to a non-peer-reviewed report describing the research (Workplace Stress Study, Spire, 2017).

Psychologists are also involved in developing a wristband called Feel that measures skin conductance, skin temperature and heart rate to recognize physiological patterns involved in emotional experience. Due for release in 2018, Feel has an associated app that will provide real-time coaching based on cognitive-behavioral therapy principles.

The first iteration of Feel is designed to help people cultivate emotional awareness and manage stress and mood. Eventually, the company hopes to develop therapeutic applications that could be delivered in real time when a person experiences distress. The goal is to recruit researchers and therapists to help pilot and test those interventions, says Feel consulting psychologist Elina Kanellopoulou, PhD.

"It is important for the company to establish collaborations with research groups who will use Feel in their studies in order to explore the potential of this technology for promoting well-being and enhancing mental health care."

Similarly, Chhabra recently designed a large trial for Sharecare, a company offering a smartphone app that measures signs of stress in the user's speech patterns during phone calls. Chhabra helped the developers refine their research questions and design a study to validate the stress ratings generated by the app.

Academics put wearables to the test

Outside of industry, independent studies have found that sensor data can predict aspects of mood and well-being. The StudentLife study at Dartmouth University passively collected automatic sensing data from 48 undergraduate students through a smartphone app. The goal was to track their mental health, academic performance and behavior trends over a 10-week period.

The project, led by Dartmouth computer science professor Andrew T. Campbell, PhD, found that students started the academic term with healthy levels of physical and social activity. Both types of activity dropped before midterms. While social activity picked up again in the second half of the term, physical activity remained low through finals (Computers in Human Behavior, Vol. 67, No. 1, 2017).

Another analysis of the StudentLife data led by Sohrab Saeb, PhD, then at Northwestern University, and colleagues focused on GPS data to assess the number of places the participants visited, how they spent time in those locations and the circadian rhythms of their movements. The researchers found they could predict participants' depressive symptoms up to 10 weeks before depression was diagnosed, particularly if they focused on patterns of weekend activity (PeerJ, published online, 2016).

At the Harvard T.H. Chan School of Public Health, biostatistics professor Jukka-Pekka Onnela, PhD, has developed a smartphone-based research platform to collect information about behavioral patterns, sleep, social interactions, physical mobility, gross motor activity, cognitive functioning and speech production. The platform, named Beiwe, for a Nordic goddess of sunlight and mental health, was built with grant money from the National Institutes of Health and is free for researchers.

Now, Onnela has teamed up with John Torous, MD, co-director of the digital psychiatry program at Beth Israel Deaconess Medical Center, to study whether Beiwe's sensor data can predict relapses in people with schizophrenia. A preliminary analysis of the data suggests that patterns of social activity and mobility could predict a pending relapse, Torous says. "We notice that before people become ill, they have changes in their own daily patterns. We're not finding a universal signal, but each person might have his or her own personal relapse signature."

Can apps change behavior?

Tracking devices could be game changers if they could reliably predict schizophrenic relapses, detect signs of depression or deliver therapy modules to curb anxiety. But designing and validating such tools is only half of the equation.

"The accuracy and validity of a device is only the very first step. If it wasn't, we would all buy bathroom scales and immediately lose weight," notes Maribeth Gandy Coleman, PhD, director of the Interactive Media Technology Center at Georgia Tech. The best stress-detector in the world won't do a bit of good if a person stops wearing it, or doesn't bother changing their stress-­inducing habits when their wristband sounds the alarm.

In some cases, the accuracy of a tracking device may matter less than how the person responds to it. If a fitness tracker reports that you've taken 10,000 steps today when you've actually taken 9,500, it probably won't make much difference to your physical health, Coleman notes. But if wearing the device inspires you to be more active in general, it's a win.

Many questions linger about who is most likely to engage with wearable devices, and how, why and when people would use them. Unfortunately, the people who could benefit the most from the devices may be the least likely to take advantage of them, Torous says. "In a lot of mental illnesses, where lack of motivation is a symptom, people may be even less engaged with the technology," he says.

Research psychologists and other behavioral health experts can help determine who buys mental health wearables or downloads the apps—and who sticks with them for the long haul. "People already own phones, and the technology is getting cheaper to build. It's not so much a technical problem anymore," Torous says. "In some ways, it's becoming a clinical research question, and we need the research community to embrace it."

Given that the technology and research are both in their infancy, some clinicians wonder what to do when patients bring them data from mental health apps. "I had a patient bring me wearable data, and it was hard to say what it meant. What is the baseline? What do variations mean? How reliable is it?" Torous says. "It's still hard to use clinically, which is too bad because there's a lot of relatively simple research that could help us understand those correlations."

Until that work is done, he adds, people who use mental health apps—and their mental health providers—should approach such data carefully. "It's as if we've opened up a magical box, and we're all so enthralled," he says. "But we're all still sorting our way through."

APA is hosting Technology, Mind and Society, an interdisciplinary conference exploring interactions between humans and technology, on April 5–7, in Washington, D.C. For more information or to register, visit https://pages.apa.org/tms.

Additional reading

The Digital Mental Health Revolution: Opportunities and Risks
Tal, A. & Torous, J. Psychiatric Rehabilitation Journal, 2017

Using Smartphones to Collect Behavioral Data in Psychological Science: Opportunities, Practical Considerations, and Challenges
Harari, G.M., Lane, N.D., Wang, R., Crosier, B.S., Campbell, A.T., & Gosling, S.D. Perspectives on Psychological Science, 2016

Mobile Technology for Mental Health Assessment
Areàn, P.A., Hoa Ly, K., & Andersson, G. Dialogues in Clinical Neuroscience, 2016

The Rise of Consumer Health Wearables: Promises and Barriers
Piwek, L., Ellis, D.A., Andrews, S., & Joinson, A. PLoS Medicine, 2016