Pilot program offers a road map for PCMH model of care

Internists who have attempted to convert to a patient-centered medical home model have quickly learned the process isn't always easy. But a pilot program found the results appeared worth it: fewer emergency visits and patients who take better care of their own health.

Becoming a patient-centered medical home (PCMH) isn't easy. Many internists may have already suspected or discovered as much, but a new report of one PCMH pilot quantifies some challenges of converting to the new model of care.

There was good news, too, from the Colorado Multipayer Patient-Centered Medical Home Pilot. Final results of the three-year program are still pending, but preliminary reports of the collaboration among 16 primary care practices and their insurers showed a significant reduction in emergency department visits and hospital admissions among the 100,000 or so participating patients.

The hurdles encountered by the program, including the reluctance of self-insuring employers to contribute and difficulty gathering data on health care cost, will need to be addressed by current and future PCMH efforts, according to the report in the September Health Affairs.

For more details, ACP Internist recently spoke with the report's authors: Marjie Harbrecht, MD, a family physician and CEO of HealthTeamWorks, in Lakewood, Colo.; and Lisa Latts, MD, FACP, an internist and principal of LML Health Solutions LLC, a quality and safety consulting organization located in Denver.

Q: For participating practices, what changes did being in the pilot involve?

A: Dr. Harbrecht: We'd measure them on several parameters. One was team-based care: Did they really start working more as a team as opposed to top-down? We found when they really split up the work and let everybody perform to the highest level of their licensure, it takes a lot of pressure off the physicians.

They started to use guidelines and protocols to drive improvement, such as spreadsheets to say, “Mrs. Jones is a diabetic. Here are all the things that the guideline says she needs to have done, and the time frames in which they need to have them done.” And they did outreach to patients who weren't coming in to make sure they got all the recommended care.

In addition, they started to also use self-management support to help patients engage in their own care and assist them in setting their own goals, and started to use motivational interviewing techniques.

Q: What effects did you see on the patients from these efforts?

A: Dr. Harbrecht: We did patient satisfaction surveys initially. [Patients] would make comments such as, “We liked our doctors and nurses, but we couldn't get in and we had difficulty getting information.” Whereas in the later surveys, they were saying, “Boy, I've really learned how to take care of myself. I feel much more empowered in reaching my health goals.” They were actually starting to repeat back some of the language that the practices were using. We'll be releasing another paper on some of the results, both on the practice measures as well as the patient surveys.

Q: Participating practices were paid under the traditional fee-for-service model, but they also received an extra monthly payment (about $4 to $8 per patient). What was that for?

A: Dr. Harbrecht: Because the plans put some additional funding on the table, through the per member/per month care coordination fees, they were able to do a lot of different things, such as extended hours, using secure e-mail and phone more frequently, to prevent patients from having to go to the emergency room. Perhaps the most important was adding care coordination/care management services to follow up with patients when needed, and assist them in setting and achieving their own health goals, particularly those with chronic conditions or complex issues.

Q: What motivated the health plans' involvement?

A: Dr. Latts: We really wanted to see a transformation in the way primary care worked. We have seen a phenomenon, as everybody has, where a very small percentage of members are responsible for a large percentage of costs. In Wellpoint's case it was something like 1% of patients responsible for 28% of costs, and 5% for 55%.

It was very important to focus on the individual who had chronic disease and high cost, but also to provide an infrastructure for a broader look at [a practice's] entire panel of patients. That, we realized, was going to take investment on our part. The practices don't have the ready revenue they need to make a significant investment in transformation.

Also, it was going to take a radical revision of thinking to incentivize the right thing to encourage the practice changes we were hoping for. We really liked the idea of a blended payment that would try to keep all the incentives and all the objectives.

Q: Your blended payment model had three parts: fee-for-service, the care coordination fee and a pay-for-performance bonus. Do you see that combination having potential as a permanent payment model?

A: Dr. Latts: If it's not the right way to go, it's in the right direction. It fixes a lot of the problems of the past. It's very complicated and has its own problems. Most notably, because of self-insured employers [choosing not to participate], physicians were paid through the blended payment model for approximately 20% of their patients. They delivered the intervention to 100% of their panel and we were only able to deliver the blended payment for 20%.

Also, it was very difficult especially in a pilot environment [for insurers] to automate the per member/per month care coordination fee, so that ended up being a manual process which was very labor-intensive on our side and very frustrating on the physician side.

Dr. Harbrecht: The blended model is a good start, especially for those just starting this journey. It balances overuse issues seen with fee-for-service alone, underuse issues with capitation alone, and adds incentives for the hard work needed to improve quality and reduce cost trends, which we didn't see in the old managed care days. However, fee-for-service was still too dominant [in the pilot]. It left practices on an acute care treadmill.

Q: What other challenges did the program encounter?

A: Dr. Harbrecht: In some places, the medical neighborhood [specialists and hospitals] was willing to play and in others they weren't. We were really pushing the practices to reach out to their specialists or hospitals to get information.

If the primary care physician was responsible for reducing cost trends, that became very difficult if you didn't even know that the patient went to the emergency room or went to the hospital, and what happened. You need to follow up to prevent readmissions and to educate folks. For instance, if they went to the ER for something that could have been treated by primary care, tell them, “Next time you've got a bad sore throat or a cold, we can do that. Here are our hours.”

Q: What could specialists do to improve the situation?

A: Dr. Harbrecht: Communicate back and forth. Use compacts to clarify our relationships: If I refer a patient, how do I clarify exactly what I need? Are you going to be taking on the patient's care, or is this a shared responsibility, or do I just need a quick opinion?

If we go back to managed care days, and say this is all on primary care to control cost and quality, but we do not bring the hospitals, specialists and other participants in on the same incentive program, we still end up with a fragmented environment.

The next step would be to start to figure out, through bundled payment or global payment or other mechanisms, [how] to incent people to coordinate care—to really start to work together. We're starting to see those models emerge through accountable care organizations or integrated communities of care. But regardless of the payment model used, it's going to be very important to really work on the culture changes. You can pretty much manipulate any payment program, if you don't believe in the heart of what we're trying to do.

Q: The pay-for-performance component of your pilot was somewhat unusual. It was based on program-wide measures of quality, so all participants received a bonus if the group as a whole hit the target. How did you decide on this system?

A: Dr. Harbrecht: We had to develop it from scratch. There was no off-the-shelf that we could find that was aligned with our selected measures. For quality, each practice was evaluated individually. For cost, the pilot was to be assessed as a whole to ensure large enough patient numbers.

Unfortunately we could not get aggregated cost data across all the plans to assess this. And, besides a few plans that were able to say, “Here are your highest-risk patients to go after and manage closely,” [practices] never really got data on cost. They didn't know for much of the pilot: What was their baseline for ER and hospital costs? How were they doing? We had to go by what the literature says that you need to do to affect [costs].

Q: How did the practices respond to the encouragement to cut costs?

A: Dr. Latts: We tend to think of physicians as the money spenders. With the power of the pen, they are ordering tests, spending money. What we saw in the pilot is that there's been a huge shift in how physicians viewed their role, and that they are still responsible for the care of the patient, but they are also stewards of the health care dollar. Where quality was equivalent or superior, where possible, they wanted to direct their patients to the lowest-cost commodity facility, [such as] laboratories or radiology. But they didn't have the data or the knowledge to do that and that was a real problem.

There's an increasing realization that there's a need for increased transparency around prices. From the Wellpoint side, we have implemented programs to try and make it transparent for patients who are the lower-cost providers. But I don't think we have taken full advantage and partnered with physicians in that effort.

Q: Your pilot concluded in April. What happens next?

A: Dr. Harbrecht: With the [Centers for Medicare and Medicaid Services] Comprehensive Primary Care Initiative, we'll see a continuation of the work done in these pilots and hopefully take the lessons learned and go to the next level. We learned so much in the initial pilots. Other groups were saying a lot of the same exact things: Data is crucial, redesign is crucial.

Everybody's got to come in. Employers, especially the self-funded employers, are starting to understand that if they can step up and really participate, not only does it give practices a higher penetration of dollars to do this work, there's some return on investment here.

A lot of the self-funded employers didn't really understand [the pilot]; they said we're already paying too much, why would we pay more? Wellpoint was able to show a 2.5% to 4.5% return on the dollars that they put in initially. No longer is this a phase or flavor of the month.