Institute For The Future

Mapping Health Information Ecologies

As humans, we normally make decisions in our heads by blending related data from various domains of our lives. We model each judgement in our minds by our intentions, activities, places, and time. e.g. I want to improve my health today, so I'll manage my diet, get some exercise, and engage in low stress activities. But our digital decision making is increasingly fragmented by the scattering of our digital information and artifacts into hundreds of discrete applications and services data stores across the web: separate health records, separate data from our diet and exercise apps and sensors, and separate data from other aspects of our lives. We  gather health information from thousands of websites and communities across the web from variously credible or incredible sources to influence our health decisions. But, we lack many mechanisms to fluidly blend and analyze our life data to make easy coherent decisions.

Physicians face similar information management challenges in applying the best, upto date  validated clinical knowledge to each patient encounter. ( patients bringing their digital health data and google searches along with them).  Scientific medical data is scattered across the web in fragmented databases,  and proprietary journal services, and academic servers.

At  the recent Health Foo camp,  hosted by Tim O'Reilly and Paul Tarini  RWJF Pioneer Group, I lead a working session to prototype mapping health information ecologies starting with the assumption that unrestricted flow of therapeutic information is to health care what nutrition, hydration, respiration are to health, and that both patients and caregivers have an explicit need to apply the -best- information (not -all- information) to any health process in precise context of person place, and activity.

In our one hour foo camp session we started to first map out flows of therapeutic information from the web and from personal sensors into people's personal health ecologies and PHRs, from scientific evidence into clinical information ecologies and EHRs, and finally between personal health ecologies and clinical information ecologies for optimized contextual care, noting critical interfaces and structural obstructions. Then we started to post identify specific APIs, data structures, and semantic structures that should be optimally open to ensure the best flows of therapeutic information.

It was a simple IFTF style sticky note exercise: the patient/person at the center, personal information posted on the left, clinical information on the right, and the web all around.

• Dark blue notes for data services and APIs ( application programming interfaces)
• Light blue notesfor apps data and APIs
• Orange notes for device APIs
• Red notes for obstacles or challenges
• Yellow notes for openness
• Purple notes for privacy and security issues.

After our hour session posted the chart on the wall, and foo campers added notes all weekend.

Here's a rather low resolution photosynth panorama  of the charts I took with my iPhone:

Health Foo Camp Map of Health Information Ecologies.

Ideally this exercise can be recreated, with much more rigor, followed by systematically clustering and mapping of the natural flows of what fellow foo camper, Ken Buetow  National Cancer Institute Associate Director for Bioinformatics and Information Technology eloquently calls 'Liquid Data'.