Dr. Barbara Walters, D.O., M.B.A., senior medical director at Dartmouth-Hitchcock Medical Center, details how participation in CMS’s physician group practice demo contributed to the construction of Dartmouth’s medical home and reimbursement model.
Although it wasn’t our original intention, participation in the CMS physician group practice (PGP) demo retrospectively allowed us to build the medical home, although we didn’t know at the time that we were doing that. Building the clinical model first was a key to the measure of our success in being able to contract with both commercial plans and Medicaid in our area. This project also became the model for reimbursement and for our contracting.
The CMS demonstration consists of 10 multi-specialty groups around the United States. It is only a fee-for-service (FFS) demonstration project in a Medicare environment. The patients are assigned to each of the group practices retrospectively, based on the preponderance of outpatient care that is delivered in that group practice. For example, if a patient has at least 51 percent of outpatient visits in our site, that patient is assigned to us and we are assigned the responsibility for the total cost of that patient’s care. Then, the total cost of the care that we spend or coordinate on the patient’s behalf is compared to the total cost of care to all other Medicare individuals in the area who don’t receive the preponderance of care from us. If we provide the care more cheaply if the rate of rise of the total cost of care is less than the rate of rise of our comparison group we are eligible for a bonus, which is 80 percent of the difference between the two groups. That is important because we used that basis when we began talking to the commercial health plans about negotiating a medical home pilot. The bonuses allocated for cost savings first, then for quality. You can get approximately 50 or 60 percent for cost savings and the rest are for very specific pay-for-performance (PFP) design quality metrics.
We had success. We are currently in year five of this initiative. We just received the draft report on year three. In year one we did achieve savings, but we didn’t meet the threshold for a bonus payment, which is 2 percent. We did achieve all of the quality metrics in year one, so we increased quality compared to benchmark. Year two we achieved savings, passed the threshold and achieved 98 percent of the quality metrics, so we received a $6.8 million bonus payout. Part of that payout was for quality. An internal analysis shows that we have also achieved savings in year three and the payment is currently being calculated. Because we participated in this program, we did receive CMS Physician Quality Reporting Initiative (PQRI) payments without having to do additional reporting every year.
Here are some highlights from the medical home build that we achieved by participating in this project. We learned and developed a better way of ICD-9 coding. For a successful medical home, Medicare and we believe that the population that you take care of does need to be risk-adjusted. As the only academic medical center in the area and some of the only subspecialists, we do have some adverse risk selection. We also transformed the role of the nurse into health coaches, previsit planners, care coordinators and outreach workers. We developed registries of our patients, beginning with a disease-focused registry. Then, it became very clear that it needed to become patient-focused. Most of our Medicare diabetics also had some other comorbidity, and rather than having 20 disease registries, we needed to have a patient-focused registry. We developed best practice care processes for chronic diseases and for prevention, and we spread them to all 48 care sites. We began doing post-discharge phone calls for our patients because many of our patients are taken care of in the communities in which they live, not necessarily by ourselves. The transfer and the handoff, as part of the care coordination, became a very important part of the medical home build so much so that when we began looking for a partner in the commercial world so we could test this model of better, more coordinated care, we already had the clinical infrastructure built.