The potential benefits of the “quantified person” are huge, making it possible to improve diagnosis, monitoring, and treatment of all sorts of diseases and chronic conditions. One of the biggest barriers to realizing these gains, however, are the front-line healthcare professionals who are expected to use this information. An excellent article by Jamie Hartford in Medical Device and Diagnostic Industry is titled “Wearables: Fad of Future?” It makes a number of important points.
The first problem is that most healthcare workers don’t have the available bandwidth to deal with all the information generated by fitness band and other wearable Health Tech devices. The article cites the Journal of General Medicine which published a survey showing that the average physician sees 2,300 patients per year. It’s impossible to monitor the raw data for even a fraction of that number of people in any meaningful way.
The other problem is that physicians have been trained from the first day of medical school to manage their patients’ health in an episodic manner. Patients only visit healthcare professionals when something is wrong. This triggers a round of diagnosis, treatment, and follow-up. If all is well, then the patient disappears until the next time a problem arises. There is a strong move toward “wellness visits,” but these simply take a snapshot of a relatively small number of biometric measures that are checked for trends in the wrong direction. But this single data point once a year is a drop in the ocean of data that is produced by devices operating around the clock, all year long.
The article quotes Young Sohn, Samsung’s president and chief strategy officer, in a speech at CES 2015: “What matters is the trend and the change and being able to have a smart algorithm that can pick up on the change.” If we’re to realize the benefits of all this data, we are going to have to rely on very intelligent software — likely residing in the Cloud — that can pull together data of different types from multiple sources. This includes the personal wearables that work 24/7, as well as the discrete data points from clinical tests and other sources. The software then has to make judgments about the data, and only bring the patient’s condition to the attention of a healthcare professional if something needs attention. This changes the normal workflow for the average physician, and requires that they trust the process. If too many false positives are flagged, or if too many problems are overlooked, the healthcare professionals will be hesitant to rely on the information that the system produces.
That’s a huge burden to place on the software developers, but there is a lot at stake.