The growing wearable sensor market is yielding ever-increasing amounts of consumer-derived digital data. These data can consist of many different physiologic measures such as heart rate and rhythm, sleep quality, brain activity, and physical activity levels. As many consumers and commercial organizations look toward using wearables to monitor medical conditions, clinicians may begin to find themselves in the role of a digital data decoder.
This will be no easy task, as a number of factors will complicate the decoding and force the clinician to become a digital “detective.” Firstly, medical settings largely rely on using technologies and equipment that have been tested and validated for medical use. In contrast, most wearables are consumer products, and while they produce digital data, this does not mean the data are reliable or valid for medical use.
Secondly, most clinicians have limited awareness and formal training in how to evaluate wearables and the data they produce. Therefore, from a knowledge perspective, clinicians are disempowered from assessing the data they are presented with.
Finally, even if we have impactful and valid consumer-derived data, we must integrate the presentation of this data into the clinician workflow. Without workflow integration, clinicians will be disempowered from using these data from a process perspective. Clinicians need a time-efficient method of storing and standardizing the data obtained from different devices.
Lagging far behind our ability to collect data is the value driver for all of our sensor-driven devices or apps: big data analytics. Powerful analytics turn “bad” data into data that can drive improvements in medical treatment, research and cost efficiency. Pharmaceutical companies are beginning to see the value of data from wearable devices and are increasingly incorporating them into clinical trials in order to better understand disease processes. Health and technology collaborations like the one between UCSF and Samsung, to create the Center for Digital Health Innovation, are important in helping to make sense of all the available digital data and devices, and in defining which are the most useful and relevant to health care.
Many questions remain unanswered: Is it time we teach medical students about wearables? How soon will we prescribe a sensor or an app with a pill? Will much big data drive the next generation of medical discoveries? I believe we should be preparing for all of these now.
Send this to a friend