By Sarah Klein
Halfway through the Physician Group Practice (PGP) Demonstration, early indicators suggest the 10 large, multi-specialty group practices participating in the three-year trial are making strides in identifying Medicare patients with chronic, high-cost conditions and closing the gaps in their care.
The physician groups that succeed in improving patient outcomes and lowering overall health care costs will share the savings with the Centers for Medicare and Medicaid Services (CMS). Toward that end, all of the projects target patients with high-cost conditions, such as diabetes, congestive heart failure, and chronic obstructive pulmonary disease, who are likely to benefit from disease and case management strategies.
Billings Clinic in Montana, for example, began monitoring its 450 congestive heart failure patients, looking for signs of weight gain, an indication of fluid retention and the need for a diuretic. Using an automated telephone survey, the clinic flagged patients who needed follow-up care or medication adjustments before their symptoms worsened—interventions that enabled them to avoid deterioration in their health status and expensive hospitalizations. Over a five-month period, the clinic has avoided 65 hospital admissions, improving care and saving the Medicare program approximately $500,000.
The Washington State–based Everett Clinic has reduced hospitalizations of dying patients by enhancing its palliative care program. (The total number of admissions for patients receiving palliative care was 1.9 over an 18-month period, compared with 2.4 for patients who didn't receive palliative care.) Everett Clinic also reduced readmissions by scheduling all elderly patients for their first follow-up visit within four to 10 days of hospital discharge. (Read the case study in this issue to learn about the experience of another participant, the Marshfield Clinic of Wisconsin.)
Feasible for Solo and Small Practices Too?
These results are typical of the projects being implemented under the Medicare demonstration. Given their apparent success, the question is whether these types of initiatives can be extended to other large physician groups and to solo and small groups, the setting where the majority of U.S. physicians practice. Will the lessons learned in large group practice settings translate to smaller practices with a narrower range of specialties, fewer support staff, and less technology?
We asked leaders of physician groups participating in the PGP Demonstration what impediments they foresee in adapting their approaches for small groups, and what solutions they propose. Most repeated this caution: small groups will be reluctant or unwilling to invest in the systems and workflow changes necessary to implement these programs without more refined data from CMS establishing the financial benefit of doing so.
"This is likely to be the primary impediment to continuation and broader adoption of the PGP Demo project financial model," says Mark J. Selna, M.D., associate chief medical officer with Geisinger Health System in Danville, Penn.
The groups now participating in the demonstration are taking a gamble that their initiatives will pay off, something small groups may not be able to do. It is difficult, if not impossible, for physician groups to monitor their financial progress because they don't know which patients are assigned to them, or anything about the control groups against which their performance will be measured.
The demonstration is based upon a retrospective analysis of fee-for-service claims. Medicare patients will be assigned to physicians groups that provide the largest share of their outpatient care. To determine cost savings, expenses for that group will be compared with a control group comprised of Medicare beneficiaries in the same geographic area adjusted to reflect the clinical profiles of the patients assigned to the practice.
To share in the cost savings, expenses for designated enrollees must be at least 2 percent below the target level established for the practice by CMS. The target is equal to the adjusted base-year costs for the patients assigned to the practice, inflated by the rate of increase in Medicare costs for the control group. (The portion of savings greater than 2 percent will be divided between CMS and the physician groups. Medicare will keep the first 20 percent and the physician groups will receive the remainder, if they meet performance and quality measures. For further details of the payment methodology, download this CMS report.)
"It's a crapshoot," says Caroline Blaum, M.D., an associate professor of internal medicine with the University of Michigan Faculty Group Practice. "We don't even know if the patient is assigned to us," Blaum says.
More Financial Data Needed
CMS will evaluate each group's performance according to both quality and cost measures. The quality measures are based on threshold and improvement targets, and include measures from the Doctor's Office Quality project being conducted by CMS' Quality Improvement Organizations as well as Health Plan Employer Data and Information Set (HEDIS) indicators.
Some participants value the experience they are gaining in a pay-for-performance system, even though they are not certain of profiting—and might even lose money—from participating in the program. "If we did everything we wanted to do, we would save Medicare $9.5 million," says Doug Carr, M.D., medical director of Billings Clinic, which focuses its disease and case management efforts on patients with congestive heart failure and diabetes. To get that result, "we would spend $4 million, and we would get back $3.5 million," he says. "The important point is that, although most of the groups had similar figures, we all decided to sign the contract with CMS because we believed that payment reform was necessary and that we had an obligation to our patients and fellow health care organizations to try to improve our current system."
To attract physicians and gain their support, the program must be profitable for their practices. "The conversion for physicians really comes from payment," says James Rogers, M.D., department chair for primary care and clinical director for the demonstration project at St. John's Health System in Springfield, Mo.
Without appropriate financial incentives, CMS will continue to discourage the very improvements it wants to encourage. Integrated systems face particular challenges, because the savings to Medicare from better coordinated physician care may be seen as a loss of revenue to the hospital.
Congestive heart failure is a good example, says David Abelson, M.D., senior vice president and chief medical information officer for Park Nicollet Health Services in St. Louis Park, Minn. "We have known for years that daily contact with a patient and daily weigh-in and a nurse who can follow protocol work to keep the patient more functional," and reduce hospitalizations, Abelson says. But doing so means higher expenses at the clinic level and lower revenues for the hospital. "The incentives are so perverse even well-intended clinics have not been able to sustain a program that is shown to be effective."
Linda Magno, director of Medicare demonstrations for CMS, understands these concerns. "It wouldn't surprise us if small group practices had difficulty with the uncertainty of the model." But she cautions that it not set in stone. The incentives can be hammered out after the approach has been proven effective. "The model is just that: a demonstration intended to test not so much whether this payment model works—in the sense of identifying patients and measuring quality after the fact—but whether changing practices changes the cost to the program and the trajectory of spending over time."
If CMS finds they do, "we can haggle about the money later," Magno says. CMS may consider structuring payment differently to eliminate the uncertainty, including using pre-payment methods, or adjusting fees for physicians who provide chronic care management.
Large Groups Recommendation: Simplicity
Program leaders were optimistic that their approaches could be replicated, even without the benefit of electronic medical records or other costly technology. "The things we have done have been very simple, and yet they are producing outcomes," says Denise Seagraves, manager of disease management for Winston-Salem, N.C.–based Forsyth Medical Group. The multi-specialty group developed color-coded disease management worksheets to prompt physicians about tests and interventions for patients with chronic diseases.
The key for many groups was to create a registry of such patients and a system for keeping tabs on their test results and appointments, as well as a mechanism for identifying treatment gaps. Some use software to identify lab results that are out of range or have a staff member follow up with patients after hospitalization to ensure they understand their new medication regimens.
Abelson says Park Nicollet has one case manager to follow its 400 patients with congestive heart failure, a job that costs $70,000 per year. That may be too expensive for a solo physician practice, but a group of four physicians would have enough patients to justify hiring a case manager if they expanded the diseases followed to include other prevalent, high-cost chronic conditions such as diabetes, high blood pressure, and other cardiovascular diseases, he says.
Smaller groups may be able to tap local resources. "Don't reinvent the wheel. Look at what is already available," says Nan Holland, R.N., M.P.H., Forsyth Medical Group's director of clinical service. Disease-specific associations such as the American Diabetes Association offer patient educational materials. Physicians may also be able to take advantage of hospital-based education programs for patients. Everett Clinic gave office space to a palliative care program to enhance that program at low expense. "Questions, consultations, and referrals are done face-to-face. It's very efficient in terms of managing communication," says James Lee, M.D., the Everett Clinic's assistant medical director.
Where to Start?
Congestive heart failure and coronary disease are good areas to start focusing on, says Barbara A. Walters, D.O., senior medical director for the Dartmouth-Hitchcock Clinic in Bedford, N.H. "The number of patients isn't overwhelming," and the impact on their care is immediately evident, she says.
Start somewhere, most leaders advised. "You need to be prepared for pay-for-performance to happen. It will happen," says Holland, of Forsyth Medical Group.
Halfway through the Physician Group Practice (PGP) Demonstration, early indicators suggest the 10 large, multi-specialty group practices participating in the three-year trial are making strides in identifying Medicare patients with chronic, high-cost conditions and closing the gaps in their care.
The physician groups that succeed in improving patient outcomes and lowering overall health care costs will share the savings with the Centers for Medicare and Medicaid Services (CMS). Toward that end, all of the projects target patients with high-cost conditions, such as diabetes, congestive heart failure, and chronic obstructive pulmonary disease, who are likely to benefit from disease and case management strategies.
Billings Clinic in Montana, for example, began monitoring its 450 congestive heart failure patients, looking for signs of weight gain, an indication of fluid retention and the need for a diuretic. Using an automated telephone survey, the clinic flagged patients who needed follow-up care or medication adjustments before their symptoms worsened—interventions that enabled them to avoid deterioration in their health status and expensive hospitalizations. Over a five-month period, the clinic has avoided 65 hospital admissions, improving care and saving the Medicare program approximately $500,000.
The Washington State–based Everett Clinic has reduced hospitalizations of dying patients by enhancing its palliative care program. (The total number of admissions for patients receiving palliative care was 1.9 over an 18-month period, compared with 2.4 for patients who didn't receive palliative care.) Everett Clinic also reduced readmissions by scheduling all elderly patients for their first follow-up visit within four to 10 days of hospital discharge. (Read the case study in this issue to learn about the experience of another participant, the Marshfield Clinic of Wisconsin.)
Feasible for Solo and Small Practices Too?
These results are typical of the projects being implemented under the Medicare demonstration. Given their apparent success, the question is whether these types of initiatives can be extended to other large physician groups and to solo and small groups, the setting where the majority of U.S. physicians practice. Will the lessons learned in large group practice settings translate to smaller practices with a narrower range of specialties, fewer support staff, and less technology?
We asked leaders of physician groups participating in the PGP Demonstration what impediments they foresee in adapting their approaches for small groups, and what solutions they propose. Most repeated this caution: small groups will be reluctant or unwilling to invest in the systems and workflow changes necessary to implement these programs without more refined data from CMS establishing the financial benefit of doing so.
"This is likely to be the primary impediment to continuation and broader adoption of the PGP Demo project financial model," says Mark J. Selna, M.D., associate chief medical officer with Geisinger Health System in Danville, Penn.
The groups now participating in the demonstration are taking a gamble that their initiatives will pay off, something small groups may not be able to do. It is difficult, if not impossible, for physician groups to monitor their financial progress because they don't know which patients are assigned to them, or anything about the control groups against which their performance will be measured.
The demonstration is based upon a retrospective analysis of fee-for-service claims. Medicare patients will be assigned to physicians groups that provide the largest share of their outpatient care. To determine cost savings, expenses for that group will be compared with a control group comprised of Medicare beneficiaries in the same geographic area adjusted to reflect the clinical profiles of the patients assigned to the practice.
To share in the cost savings, expenses for designated enrollees must be at least 2 percent below the target level established for the practice by CMS. The target is equal to the adjusted base-year costs for the patients assigned to the practice, inflated by the rate of increase in Medicare costs for the control group. (The portion of savings greater than 2 percent will be divided between CMS and the physician groups. Medicare will keep the first 20 percent and the physician groups will receive the remainder, if they meet performance and quality measures. For further details of the payment methodology, download this CMS report.)
"It's a crapshoot," says Caroline Blaum, M.D., an associate professor of internal medicine with the University of Michigan Faculty Group Practice. "We don't even know if the patient is assigned to us," Blaum says.
More Financial Data Needed
CMS will evaluate each group's performance according to both quality and cost measures. The quality measures are based on threshold and improvement targets, and include measures from the Doctor's Office Quality project being conducted by CMS' Quality Improvement Organizations as well as Health Plan Employer Data and Information Set (HEDIS) indicators.
Some participants value the experience they are gaining in a pay-for-performance system, even though they are not certain of profiting—and might even lose money—from participating in the program. "If we did everything we wanted to do, we would save Medicare $9.5 million," says Doug Carr, M.D., medical director of Billings Clinic, which focuses its disease and case management efforts on patients with congestive heart failure and diabetes. To get that result, "we would spend $4 million, and we would get back $3.5 million," he says. "The important point is that, although most of the groups had similar figures, we all decided to sign the contract with CMS because we believed that payment reform was necessary and that we had an obligation to our patients and fellow health care organizations to try to improve our current system."
To attract physicians and gain their support, the program must be profitable for their practices. "The conversion for physicians really comes from payment," says James Rogers, M.D., department chair for primary care and clinical director for the demonstration project at St. John's Health System in Springfield, Mo.
Without appropriate financial incentives, CMS will continue to discourage the very improvements it wants to encourage. Integrated systems face particular challenges, because the savings to Medicare from better coordinated physician care may be seen as a loss of revenue to the hospital.
Congestive heart failure is a good example, says David Abelson, M.D., senior vice president and chief medical information officer for Park Nicollet Health Services in St. Louis Park, Minn. "We have known for years that daily contact with a patient and daily weigh-in and a nurse who can follow protocol work to keep the patient more functional," and reduce hospitalizations, Abelson says. But doing so means higher expenses at the clinic level and lower revenues for the hospital. "The incentives are so perverse even well-intended clinics have not been able to sustain a program that is shown to be effective."
Linda Magno, director of Medicare demonstrations for CMS, understands these concerns. "It wouldn't surprise us if small group practices had difficulty with the uncertainty of the model." But she cautions that it not set in stone. The incentives can be hammered out after the approach has been proven effective. "The model is just that: a demonstration intended to test not so much whether this payment model works—in the sense of identifying patients and measuring quality after the fact—but whether changing practices changes the cost to the program and the trajectory of spending over time."
If CMS finds they do, "we can haggle about the money later," Magno says. CMS may consider structuring payment differently to eliminate the uncertainty, including using pre-payment methods, or adjusting fees for physicians who provide chronic care management.
Large Groups Recommendation: Simplicity
Program leaders were optimistic that their approaches could be replicated, even without the benefit of electronic medical records or other costly technology. "The things we have done have been very simple, and yet they are producing outcomes," says Denise Seagraves, manager of disease management for Winston-Salem, N.C.–based Forsyth Medical Group. The multi-specialty group developed color-coded disease management worksheets to prompt physicians about tests and interventions for patients with chronic diseases.
The key for many groups was to create a registry of such patients and a system for keeping tabs on their test results and appointments, as well as a mechanism for identifying treatment gaps. Some use software to identify lab results that are out of range or have a staff member follow up with patients after hospitalization to ensure they understand their new medication regimens.
Abelson says Park Nicollet has one case manager to follow its 400 patients with congestive heart failure, a job that costs $70,000 per year. That may be too expensive for a solo physician practice, but a group of four physicians would have enough patients to justify hiring a case manager if they expanded the diseases followed to include other prevalent, high-cost chronic conditions such as diabetes, high blood pressure, and other cardiovascular diseases, he says.
Smaller groups may be able to tap local resources. "Don't reinvent the wheel. Look at what is already available," says Nan Holland, R.N., M.P.H., Forsyth Medical Group's director of clinical service. Disease-specific associations such as the American Diabetes Association offer patient educational materials. Physicians may also be able to take advantage of hospital-based education programs for patients. Everett Clinic gave office space to a palliative care program to enhance that program at low expense. "Questions, consultations, and referrals are done face-to-face. It's very efficient in terms of managing communication," says James Lee, M.D., the Everett Clinic's assistant medical director.
Where to Start?
Congestive heart failure and coronary disease are good areas to start focusing on, says Barbara A. Walters, D.O., senior medical director for the Dartmouth-Hitchcock Clinic in Bedford, N.H. "The number of patients isn't overwhelming," and the impact on their care is immediately evident, she says.
Start somewhere, most leaders advised. "You need to be prepared for pay-for-performance to happen. It will happen," says Holland, of Forsyth Medical Group.