The Model Cell: Follow Dr Toussaint’s Prescription

John Toussaint, former CEO of the ThedaCare health system and current CEO of Thedacare Center for Healthcare Value, has a recent piece in The Harvard Business Review that distills much of his thoughtful advice and experience redesigning management systems in healthcare.

What should an organization do that wants to follow ThedaCare’s application of the Toyota Production System?

Start with building a model cell.

The model cell aims to demonstrate, in an integrated way, the management practices that are designed to reliably deliver excellent care. People throughout the organization can learn from a local example that has sufficient complexity to demonstrate real success and reveal real challenges.

The Model for Improvement provides the framework to design a model cell and operate it. Clarity of aims, specification of performance measures, and a catalog of change ideas linked to multiple Plan-Do-Study-Act cycles provide the foundation to construct the model cell.

The “scale of a test” advice from colleagues at API also applies to thinking about a model cell:
 

(Table 7.1 G. Langley et al. (2009), The Improvement Guide, 2nd edition, Jossey-Bass, San Francisco © Associates in Process Improvement, discussed in an earlier blog post.)
 

A Large-scale Test?

The development of a model cell sketched in the HBR article looks like a Large-scale test.

John’s first example at ThedaCare began with six months study by two nurses (full-time) and a physician and pharmacist (half-time). Subsequent design cycles engaged over 100 other staff.

That’s a lot of work.

Large-scale tests, according to the API table, have the greatest chance to deliver predicted results with strong commitment by the organization, strong degree of belief that the change will work, and low cost of failure.

From the point of view of a large integrated health system, however, the development of a model cell on the scale of ThedaCare’s initial effort might be viewed as a relatively small-scale test.

Here’s why.

When I think about about testing on a small scale, I usually ask: What is the smallest informative test that can advance our understanding?

For the model cell application described by Toussaint, the test is really about organizing people and resources in a system of management to produce reliable results.

While testing individual components (e.g. daily huddles or an improved way to flow patients through a care cycle) can build skill, work on components is not enough. In a system, it is not just the components but the relationships among the components that matter.

When you are ready to follow John’s advice, think about the smallest informative test.   Target the smallest work unit that can demonstrate the components and relationships in the management system you want to develop.