Harold Dash, MD, a cardiologist, is president of the Everett (Wash.) Clinic, a group practice with more than 300 physicians. Here Dr. Dash discusses Everett's participation in the Medicare Physician Group Practice Demonstration Project, the model for the accountable care organization, a new system that bundles charges for hospitals, physician and other providers.
Question: What was the goal of the Medicare Physician Group Practice Demonstration?
Harold Dash: The goal was to lower costs and improve the quality of patient care. It involved 10 large physician groups across the country serving a total of 223,203 Medicare fee-for-service patients. Some of the practices are freestanding like Everett and some are integrated into health systems. The project lasted five years, ending this spring. Participants received the usual fee-for-service payments and then were eligible for cost performance payments and quality performance payments.
Q: How were the cost performance payments calculated?
HD: Participating groups were awarded payments for savings they generated, as compared with charges of a control group in the community. To qualify for these payments, a group's savings had to exceed the target expenditure level by 2 percent. Of that amount above 2 percent, 80 percent went to the practices and Medicare retained the remaining 20 percent.
Q: How much did participants earn?
HD: All groups received payments for improved quality measures but only four groups received performance payments. Everett was one of those groups, but it only received a relatively small payment in the second year. [For cost performance in the second year, Dartmouth-Hitchcock Clinic received $6.69 million, Marshfield Clinic received $5.78 million, the University of Michigan Faculty Group Practice received $1.24 million and Everett received $129,268. Added to $127,021 it received for quality measures, Everett received a total of $256,289 in the second year. Payments for the full five years have not been calculated yet.]
Q: What prevented Everett from making more money?
HD: We are in a very low-cost area in Washington State to begin with, so it was difficult to bend the savings curve. Also, we were not coding properly for severity of illness when there were two diagnoses, such as complicated diabetes with kidney disease. But we've learned a lot in the last 1½ years.
Q: What measures did Everett take in the five-year program?
HD: A nurse coach met in person with patients before discharge to review instructions with them and make follow-up physician appointments. Primary care physicians followed up with patients within five days of hospital discharge to address unsolved or new healthcare problems. Palliative care programs were deployed in physicians’ offices to improve end-of-life care for 800 patients. Primary care physician physicians received electronic patient reports on diabetes, heart disease, hypertension, as well as mammogram and colonoscopy screening results. We implemented evidence-based guidelines for ordering imaging tests.
Q: Do you see any flaws in the demonstration program?
HD: It presented a number of challenges. There was no credit for upfront infrastructure costs and it cost Everett more than $1 million to put together the program. Savings had to exceed the 2 percent threshold before groups could be paid, which meant participants did not receive a good deal of the money they saved. [In the project's second year, for example, all 10 participants saved Medicare a total of $34 million for participating beneficiaries, but CMS recognized only $17 million in savings due to the 2 percent threshold.]
There were no aligned incentives between the Everett Clinic and other providers, such as nursing homes, and we didn't know who our patients were. It was not clear which of our patients were in the program. Also, the program was very slow to provide feedback data. We didn’t receive data back until a year later. It's hard to act on information when it comes that late.
Q: Is Everett planning to launch an ACO?
HD: We all would love to participate in an ACO, but it needs to be practical. I don't think any of us know exactly what an ACO is going to be. All we have so far are the bare bones of the program as stated in the healthcare reform law. We need to see the regulations. Also, there may be practical limitations not mentioned in the bill. For example, the bill says an ACO needs a minimum of 5,000 patients to run an ACO, but we believe it probably needs 10 times that amount to be financially feasible.
Q: Who is best positioned to run an ACO, a group practice or a hospital?
HD: Hospitals have different incentives than physicians. They have to fill their beds. On the other hand, hospitals are beginning to control a large number of group practices. That may change the dynamic. Also, starting an ACO is expensive and group practices generally don't have access to funds because earnings are distributed every year. Everett is an exception, however. Our goal is to put 5 percent of earnings back into the organization every year. We've achieved this for most years but it's going to be a challenge this year.
Learn more about the Everett Clinic.
Question: What was the goal of the Medicare Physician Group Practice Demonstration?
Harold Dash: The goal was to lower costs and improve the quality of patient care. It involved 10 large physician groups across the country serving a total of 223,203 Medicare fee-for-service patients. Some of the practices are freestanding like Everett and some are integrated into health systems. The project lasted five years, ending this spring. Participants received the usual fee-for-service payments and then were eligible for cost performance payments and quality performance payments.
Q: How were the cost performance payments calculated?
HD: Participating groups were awarded payments for savings they generated, as compared with charges of a control group in the community. To qualify for these payments, a group's savings had to exceed the target expenditure level by 2 percent. Of that amount above 2 percent, 80 percent went to the practices and Medicare retained the remaining 20 percent.
Q: How much did participants earn?
HD: All groups received payments for improved quality measures but only four groups received performance payments. Everett was one of those groups, but it only received a relatively small payment in the second year. [For cost performance in the second year, Dartmouth-Hitchcock Clinic received $6.69 million, Marshfield Clinic received $5.78 million, the University of Michigan Faculty Group Practice received $1.24 million and Everett received $129,268. Added to $127,021 it received for quality measures, Everett received a total of $256,289 in the second year. Payments for the full five years have not been calculated yet.]
Q: What prevented Everett from making more money?
HD: We are in a very low-cost area in Washington State to begin with, so it was difficult to bend the savings curve. Also, we were not coding properly for severity of illness when there were two diagnoses, such as complicated diabetes with kidney disease. But we've learned a lot in the last 1½ years.
Q: What measures did Everett take in the five-year program?
HD: A nurse coach met in person with patients before discharge to review instructions with them and make follow-up physician appointments. Primary care physicians followed up with patients within five days of hospital discharge to address unsolved or new healthcare problems. Palliative care programs were deployed in physicians’ offices to improve end-of-life care for 800 patients. Primary care physician physicians received electronic patient reports on diabetes, heart disease, hypertension, as well as mammogram and colonoscopy screening results. We implemented evidence-based guidelines for ordering imaging tests.
Q: Do you see any flaws in the demonstration program?
HD: It presented a number of challenges. There was no credit for upfront infrastructure costs and it cost Everett more than $1 million to put together the program. Savings had to exceed the 2 percent threshold before groups could be paid, which meant participants did not receive a good deal of the money they saved. [In the project's second year, for example, all 10 participants saved Medicare a total of $34 million for participating beneficiaries, but CMS recognized only $17 million in savings due to the 2 percent threshold.]
There were no aligned incentives between the Everett Clinic and other providers, such as nursing homes, and we didn't know who our patients were. It was not clear which of our patients were in the program. Also, the program was very slow to provide feedback data. We didn’t receive data back until a year later. It's hard to act on information when it comes that late.
Q: Is Everett planning to launch an ACO?
HD: We all would love to participate in an ACO, but it needs to be practical. I don't think any of us know exactly what an ACO is going to be. All we have so far are the bare bones of the program as stated in the healthcare reform law. We need to see the regulations. Also, there may be practical limitations not mentioned in the bill. For example, the bill says an ACO needs a minimum of 5,000 patients to run an ACO, but we believe it probably needs 10 times that amount to be financially feasible.
Q: Who is best positioned to run an ACO, a group practice or a hospital?
HD: Hospitals have different incentives than physicians. They have to fill their beds. On the other hand, hospitals are beginning to control a large number of group practices. That may change the dynamic. Also, starting an ACO is expensive and group practices generally don't have access to funds because earnings are distributed every year. Everett is an exception, however. Our goal is to put 5 percent of earnings back into the organization every year. We've achieved this for most years but it's going to be a challenge this year.
Learn more about the Everett Clinic.