When Will Digital Health Concepts Reach the Doctor?

The always insightful Joseph Kvedar, MD, has a great post up where Joe gets a wake-up call from one of his advisors:

“Joe, when are we going to take all of these digital health concepts from the 30,000 foot level and get them into that 10 minute window that the doctor has with the patient?” It is not hyperbole to say that this put the last 20+ years of my career in a whole different perspective.

This is a good wake-up call for all of us in the space. Pushing digital health solutions down to the 10 min window a doctor has with a patient is the nirvana of what we’re doing and is incredibly challenging.

Dr. Kvedar suggested that we’ve already started to do this when he shared an example of how his PCP offered an eVisit option for follow-up to his in-person visit. I think that is a good example, but his insights into the 2nd phase offered a great look into where all of this is headed [emphasis added]:

Phase two will be the integration of tools like remote monitoring of diabetes and blood pressure. This is more tricky. The front-end work of monitoring remotely-derived values is done by either a non-physician clinician or, in some cases, a software algorithm. The doctor gets involved only when there is a complex medical judgment required. When deployed at scale, this approach extends the doctor across many more patients due to the one-to-many nature of the intervention.

Taking the recent interaction with my PCP as an example, remote monitoring would be considered a whole new channel of work, which doesn’t easily fit in to his workflow like an evisit does. It is hard to estimate its value, hard to predict how much impact it will have and hard to envision how to integrate it into clinical practice.

There are some real gems in this quote. My favorite is “The doctor gets involved only when there is a complex medical judgment required.” The future of healthcare IT is going to be built on this concept. When does a doctor need to get involved and when can another staff member or software algorithm address the situation? It will take us at least a decade or more to figure out this balance. Not to mention the workflow that will make sense.