The Work of Transforming Primary Care

May 1, 2014 | Christopher F. Koller, President

I recently addressed a gathering of about 50 health leaders—providers, payers and state officials—who had come together at a Milbank Memorial Fund-convened meeting as part of their work on implementing local multi-payer primary care enhancement projects.

This is just the kind of work the Fund likes to support—improving population health by linking leaders and decision makers with the best evidence and experience. The evidence on the value of primary care is strong—and the projects underway are beginning to generate some significant positive outcomes.

Sixteen states were represented in the room. Collectively, the projects touched 1,400 practices, 7,800 providers, and 3.75 million patients. Among the participants were projects participating in the Medicare Advanced Primary Care Practice (MAPCP) demonstration, in the Comprehensive Primary Care Initiative (CPCI), and those with no Medicare participation.

In my address, I asked the people in the group—people who were toiling in the fields of their respective states, powered more by passion than by resources, drawing slingshots at the Goliath-like status quo—this question: If you were asked to describe this work, what would you say?

When I was Health Insurance Commissioner in Rhode Island doing this same kind of work, I probably would have answered by saying something like – “We are developing and implementing a multi-payer pilot program to implement the patient-centered medical home.” Which is probably the way many of the people in the room would have answered. But there is much more to it than this rather dull explanation.

Let me explain by taking a step back. In creating our invitation list, we tried to identify every project around the country that had the following eight components:

  1. Innovative payment reforms to support primary care
  2. Multiple payer participation
  3. State government playing a convening role
  4. Standards for patient-centered medical home identification
  5. New staffing models for team-based primary care
  6. Technical assistance to practice sites
  7. Common measurement of performance
  8. Collaborative learning

Doing these eight things well takes enormous amounts of work—and yet we identified 16 different projects that were taking it all on. Examining the projects, we found there were wide variations in emphasis on these components and in their design. Articulated this way, the projects seemed like a set of complex technical and design challenges to change systems. And they were.

In talking to state health policymakers about their work, however, we discovered that it takes much more than a set of technical challenges to design the right payment mechanism or the right staffing mix to implement a patient-centered medical home. This, it turns out, might be the easy part.

Suppose the people who do this work described it this way – “we are transforming the practice of primary care in our communities.” I think that is a more accurate description.

Put this way, this work is more than a technical challenge, it is a social one. These states are trying to change institutions, cultures and individuals from long-held ways of doing and being, and creating systems that are consistent with—not willfully ignorant of—the evidence of the value of primary care for population health.

Now, if we were dealing with commodities and services, we could let the market do the dirty work of spreading and adapting this innovation of the patient-centered medical home – just as the market has spread the innovation of the smart phone, changing entirely the way we think of taking pictures.

But we are not talking about a commodity and we cannot rely on the market to transform primary care. Individual shoppers don’t seek out and reward good population health—it remains a public good. A challenge for these projects was to make measuring and improving population health a priority for medical care providers, payers, and patients in the same way that accurate diagnoses, effective treatments, revenue enhancements, and expense minimization already are. This is foundational. If you are about transforming primary care in your communities, you are unavoidably in the business of social change—changing people’s attitudes and public policies as well as changing systems. You have to do both.

Foundations, because they are drawn to big social problems, have latched onto the idea of “collective impact”—the idea that social problems require collective action and that positive impact is possible. The 5 conditions for collective impact consist of:

  1. Many participating organizations – one problem; one agenda for change.
  2. A shared measurement system with a common definition of success.
  3. Mutually reinforcing activities by participants to carry out that agenda.
  4. Continuous communications over a long period to inform and build trust.
  5. An independent backbone organization for ongoing support.1

If the work of these states is both systems change and social change, then they have to accomplish this second set of tasks as well the first. They have to ask if they really have assembled all the elements on this second list, as well as the first one. Are the right people at the table? Do they define the agenda the same way? Are they measuring the same stuff the same way and using those measures? Are their actions consistent with the agenda? Leaders have to understand their own role: do they lead one of the participating organizations or are they perhaps the entire “backbone organization”? Are they trying to do both?

Given the challenge each project faces of facilitating systems change and social change, the Fund’s purpose in convening them was very simple: to help them get better faster. Payers will not wait forever for results. Stakeholder attention will drift to something brighter and shinier—Accountable Care Organizations (ACOs), “big data,” “consumerism.” The key to quicker improvement is measurement. When I was in Rhode Island, we could show improvements in the quality areas we selected but not in the utilization. And the reasons were simple—our measures in quality were better; they were more frequent, more reliable and more widely distributed. At this meeting, the group was going to share the measures they had—incomplete and inconsistent as they might be—and learn from them. Learning requires collegiality. It is relative and indefinite, but the greater risk is in inaction and lowered expectations. This process is like sailing—one of constant corrections based on the feedback one is getting and a clear sense of the destination.

As these projects develop, especially the MAPCP projects, and we begin to get some results, we are learning some exciting things:

  • Multi-payer primary care transformation is in fact working. We can demonstrate that the interventions are moving us closer to the Triple Aim of better population health, lower costs, and improved patient experiences.
  • It is working better in some places than in others
    • Yes the data is early and directi
      onal, but we are starting to see differential performance.
  • The clock is ticking for participation, especially for Medicare.
    • We cannot rely on passion to justify what we are asking of others—we must develop the capacity to learn and improve.

There are easier ways to make a living than trying to facilitate systems change and social change; institutions are many and interests are somewhat entrenched. But these projects are showing that change is possible. The evidence on the value of primary care is irrefutable, and the people in that room were taking the steps necessary to transform primary care and improve the health of the people in their communities. And that, I told them, is what you can tell others you do.


1 Kania, John and Kramer, Mark. “Collective Impact.Stanford Social Innovation Review, Winter 2011. p. 36-41.