To better understand the link between poor mental and physical health, Benjamin W. Nelson, PhD, a clinical research scientist at the online mental health provider Meru Health and a postdoc at the University of North Carolina at Chapel Hill, is studying the ability of common wearable devices like Fitbits and Apple Watches to capture relationships between physiological patient variables such as cardiovascular health and physical activity, and psychological variables like patients’ mood states.
The idea came from Nelson’s clinical practice days, when he was seeing several patients who reported greater daily fluctuations in mood and behavior over the course of the week than he was seeing in cross-sectional and longitudinal lab research designs that collected data every few months.
“I started to wonder if our infrequent, controlled laboratory environment settings were actually generalizing to people’s real-world daily experiences,” said Nelson.
That curiosity inspired him to launch several studies focused on capturing patients’ daily experiences via smartwatches and fitness-monitoring technology. After demonstrating that such commercial devices could produce accurate physiological data, he’s gone on to show differences in step count, sleep duration, and other measures among teens with varying levels of anxiety and depressive symptoms, for example (Journal of Child Psychology and Psychiatry, in press).
With enough research, Nelson hopes such tools will be available to therapists—with appropriate consent from patients—to better understand patterns of mood and physiology. For example, device data could partially supplant the 15 minutes of homework reviews and recalling of the past week at the start of a session by objectively showing how a person actually did over the week, thus freeing up valuable therapy time. Eventually, such data could also be used to determine when a client might be in serious trouble, he added.
But the area is still in its infancy, and much more research needs to be done before such data could or should be used in practice, said Nelson, who along with colleagues recently published guidelines on the use of wearables in research (NPJ Digital Medicineopens in new window,opens in new window online, June 26, 2020).
Among their recommendations and concerns:
- The field needs uniform research standards on how to collect and use data tracked by wearables. Commercial devices are proprietary, and companies can change algorithms any time they want to, in ways that may quickly render research findings inaccurate. “So, we need clinicians, academic researchers, and industry to work together to ensure that any research results are optimally safe and accurate,” Nelson said.
- Studies must be large enough to capture demographic differences and address issues related to health disparities. For example, research shows that the amount of melanin in the skin—its level of pigmentation—can affect the accuracy of some heart-rate measures, so both users and clinicians need to be familiar with these types of distinctions when interpreting data. And because wearable devices are used most often by affluent individuals, there’s concern that they will be overrepresented and that other groups will be underrepresented.
“There’s great promise to a lot of these technologies,” Nelson said, “but deploying any of them prematurely could result in negative effects that we didn’t necessarily foresee.”
