What origami can teach us about population health management

By Pranam Ben

My daughter and I were recently learning about origami, the ancient Japanese art of paper folding. As we were working on designs together, I found myself becoming very frustrated. Not only was folding the little pieces of paper challenging, but envisioning how to achieve the end result was much more difficult than I imagined.

My natural reaction when faced with anything new and challenging is to learn more. After a few Google searches I discovered Robert J. Lang, PhD, an American physicist who is also considered the finest origami artist in the United States. The secret behind Lang’s brilliant origami designs is quite applicable to any field of creative arts, sciences, engineering or business. That secret? He says: “Almost all innovation happens by making connections between fields that other people don’t realize.”

Lang’s quote perfectly aligns with what we do at The Garage. Helping ACOs and other clinically integrated networks (CINs) find the hidden connections between data from multiple sources and generating previously unknown intelligence is what drives all of our platform features. Exposing those hidden connections, however, is just the first step. CINs must act on those newly discovered trends to help patients engage in their care and make healthier decisions.

Transitioning to a Clinically Intelligent Network

CINs that are leveraging those connections for insight-driven interventions are what I like to call clinically intelligent networks. They are using analytics and automated workflows to more efficiently and effectively manage patient populations.

Transitioning to a clinically intelligent network involves answering four key questions for every patient and population:

1. What happened?  Capturing and sharing historical data is the first step in becoming an intelligent network, but discovering the underlying causes needs to follow.
2. Why did that happen?  Advanced analytics technology available today, such as our Bridge platform, is how we start to make the connections that are unseen by the naked eye. In addition to clinical data, it incorporates demographic, environmental, lifestyle and many other data points to accurately determine why health trends are deteriorating or improving.
3. What will happen?  Advanced technology powered by machine learning capabilities is helping predict care needs. CINs are intervening earlier and much more accurately around key costly events such as the likelihood of an ER visit or admission, care-plan adherence and usage of preferred network providers.
4. What should I do?  With these predictions, CINs can set up automated notifications and artificial intelligence (AI), and initiate outreach activities, such as text messages or mobile surveys, to help patients stay on track with their care plan. Frontline clinician workflows can be aligned with the AI functionality, so they can concentrate on the most challenging patients and help them stay adherent to their care plan.

High-performing CINs that have explored these four questions are seeing tangible results in terms of predicting emergency department utilization, managing referral networks and forming spending forecasts that have been accurate to only a 0.25-percent variance.

In the coming months, we’ll describe in greater detail how machine learning and AI is helping CINs discover hidden connections and helping providers engage more patients in less time and with less effort. In the meantime, please read my recent article in ACO News here to learn more about the four steps to exposing and acting on hidden population health management connections.