Issue: Improving the quality and efficiency of health care for patients with chronic illnesses requires continued investment in infrastructure such as information systems and care management processes. However, physicians are not financially rewarded for such initiatives under Medicare's current fee-for-service payment schedule, and they may even suffer reduced revenues after making these investments—both of which act as disincentives to improvement. [1]
In response to this concern, Congress in 2000 directed the federal government to conduct a three-year Physician Group Practice (PGP) Demonstration to test how physician payment incentives might promote greater coordination of care, more efficient service delivery, and improved health outcomes for Medicare beneficiaries.
The Centers for Medicare and Medicaid Services (CMS) selected 10 large physician groups in 2005 to participate in Medicare's first pay-for-performance initiative for physicians. This case study examines the early experience of one participating site, the Marshfield Clinic.
Objective: The Marshfield Clinic's mission is "to serve patients through accessible, high-quality health care, research, and education." Strategically, the clinic seeks to achieve the Institute of Medicine's six aims for a high-performance health system: safe, timely, effective, patient-centered, efficient, and equitable health care. The clinic joined the PGP Demonstration to learn how performance incentives could advance these goals, in particular by leveraging information systems to support a population health approach to patient care.
Organization and Project Leadership: The Marshfield Clinic, founded in 1916, is a multi-specialty group practice. Its 730 physicians and 6,000 staff serve more than 360,000 patients at 41 ambulatory care sites located in 35 Wisconsin communities. The physician-governed, nonprofit organization serves all who seek care, regardless of their ability to pay.
Theodore Praxel, M.D., M.M.M., Marshfield Clinic's medical director for quality improvement and care management, and Marilyn Follen, R.N., M.S.N., administrator of quality improvement and care management, oversee the clinic's participation in the PGP Demonstration.
Implementation Timeline: The Medicare PGP Demonstration began on April 1, 2005, and will run through March 31, 2008.
The Marshfield Clinic has electronically coded clinical information on all patients dating back to 1960 and has operated an electronic medical record (EMR) of growing sophistication since 1985. The EMR captures patients' diagnoses, procedures, medications, test results, radiology images, and physicians' notes. Clinic physicians began using wireless, tablet-style personal computers for electronic prescribing and dictation in 2003. The clinic plans to completely eliminate the use of paper charts (a process already underway) by the end of 2007, when every physician is scheduled to have a tablet PC.
Target Population: Elderly and disabled fee-for-service Medicare beneficiaries are assigned to the demonstration if they receive the plurality of their outpatient care from a participating physician group.
Key Measures: The PGP demonstration uses 32 performance measures, a subset of those developed by the American Medical Association's Physician Consortium for Performance Improvement for the Medicare Doctors' Office Quality project being conducted by Medicare's Quality Improvement Organizations. Measures are added incrementally over the three-year demonstration, as shown in the Table below.
A participating physician group may earn a bonus of up to 80 percent of any Medicare cost-savings that it achieves that exceed 2 percent of its expenditure target (the group is not penalized if it does not meet its target). The expenditure target is based on the practice's own base-year costs inflated by the risk-adjusted annual expenditure growth rate for a comparison group of Medicare beneficiaries. If the PGP qualifies for a bonus, a portion (30 percent the first year and rising to 50 percent by the third year) is tied to the physician group's performance on quality targets. Medicare retains the remaining 20 percent of savings achieved by the PGP plus any bonus set aside for quality performance that is not earned by the PGP. [2]
Process of Change: Marshfield Clinic is using electronic tools to improve workflow and promote quality improvement in several ways. Ongoing information technology investments, including the cost of its electronic medical record (EMR), comprise about 3.5 percent of its annual revenue.[3]
The Marshfield Clinic is redesigning its appointment scheduling process to increase the accessibility and continuity of primary care. [4] By reserving some appointments for daily needs, patients are more likely to see their usual physician in a timely manner when they need care. This approach may help prevent complications and more costly care for those with chronic conditions such as heart failure. It also eliminates duplicate visits for those who would otherwise see another physician or visit the emergency room and then require follow-up with their regular doctor.
The clinic seeks to involve physicians in quality improvement through several mechanisms, including online guidelines and storyboards, continuing medical education, and coaching by four regional quality-improvement medical directors. The medical directors attend departmental meetings to share performance results and improvement strategies and solicit feedback. Local sites are becoming engaged in redesign efforts that aim to optimize workflows. For example, medical assistants have been trained to conduct diabetic foot exams, with physicians becoming involved only when there is an abnormal result requiring intervention.
Early Results:
A well-developed electronic medical record system can support performance goals by improving physician productivity and patient care, thereby creating clinician demand for EMRs. (Marshfield Clinic clinicians are currently on a waiting list to receive their tablet PCs.) When asked if there is physician resistance to the EMR, Praxel responded, "that is like asking me who [doesn't want to] use electricity or water. It's become an integral part of the practice."
On the other hand, Praxel cautions that EMR adoption should not viewed as a solution in and of itself. "It's a good tool; but one has to take the data and turn it into actionable information." In particular, the clinic is seeking to leverage IT to promote population health through the feedback of performance data to physicians that enables them "to examine [their] practice and take action to improve," says Follen.
"Physicians in general are used to thinking about patients one at a time," Praxel explains. "While we don't ever want to lose the individual aspect of patient care, we're trying to close the loop from the individual back to the population to raise awareness of how the physician is doing with a given disease or constellation of diseases for their entire panel of patients." Physicians can then apply that broader perspective to improve the care of individual patients.
Despite the emphasis on information technology, Praxel sees the clinic's quality improvement process as primarily non-technical in nature. "We're using IT to help create transformational change in how we deal with patients," he says. "That's why we have regional quality improvement directors meeting with departments, explaining why it's necessary to alter the way care is currently delivered to patients, and getting physicians involved in change as it's occurring," he says.
Follen believes that an internally developed care management program fosters physician acceptance of change. "Involve your providers so it is truly an extension of their practice and so patients don't feel that they are being dismissed by the physician and managed by a nurse," she says. By integrating care management into clinic operations, "everyone feels it is a team effort to improve care for the patient and the population."
Moreover, process improvements must be tailored to the local environment rather than a forcing a "one-size-fits-all" approach. For example, what is optimal in a large clinical center may be very different from what is optimal in a small one, Praxel points out.
Implications: Although an internal systems development process was well suited to Marshfield Clinic's circumstances, it may not be the appropriate path for the smaller physician groups in which most physicians practice and most patients receive their care. Praxel, who spent 10 years working in a group of 10 to 12 providers before joining the Marshfield Clinic, isn't confident that many small groups have the time and resources necessary to duplicate the clinic's homegrown infrastructure. Other strategies to adapt and embed off-the-shelf systems might be an option for smaller practices.
The capitation-like incentive structure created by the PGP Demonstration may not be broadly applicable to physician practice outside large groups. For example, the Marshfield Clinic does not own or control hospital facilities and faces barriers in proactively coordinating inpatient care because of legal concerns about sharing patient information between separate entities. This may put the clinic at a disadvantage in controlling global costs in comparison to fully integrated PGPs (that include both physician groups and hospitals) participating in the demonstration. It also highlights the potential limits of pay-for-performance in achieving transformational system change.
Establishing common definitions will be critical to furthering the adoption of information technology and performance measurement. The clinic, says Praxel, found "different definitions for monitoring quality from different payers to be taxing and costly." The Institute of Medicine recently recommended that the federal government establish a standards body to address this concern. But, until electronic systems are interoperable, there will continue to be wasteful duplication and gaps in information as people move from one system to another.
For Further Information: Contact Theodore Praxel at [email protected] or Marilyn Follen at [email protected].
References
[1] L. Casalino et al. (2003) External Incentives, Information Technology, and Organized Processes to Improve Health Care Quality for Patients with Chronic Diseases. Journal of the American Medical Association 289, 434–441; A. J. Audet et al. (2005) Physicians' Views on Quality of Care: Findings from the Commonwealth Fund National Survey of Physicians and Quality of Care. New York: The Commonwealth Fund.
[2] RTI International (2005) Physician Group Practice Demonstration Quality Measurement and Reporting Specifications, Version 2. Baltimore: Centers for Medicare & Medicaid Services.
[3] Marshfield Clinic (2003) Integrated Computer Technologies at Heart of Ability to Care for Patients Effectively. Cattails (Nov/Dec).
[4] M. Murray and D. Berwick (2003) Advanced Access: Reducing Waiting and Delays in Primary Care. Journal of the American Medical Association 289, 1035–1040.
[5] M. Hillman (2002) Testimony Before the Subcommittee on Health of the House Committee on Ways and Means, Hearing on Promoting Disease Management in Medicare (April 16). Washington, D.C.: U.S. House of Representatives.
[6] Marshfield Clinic—Indianhead Center (2006) Improving Access in Primary Care—Virtually. Improvement Report. Boston, Mass.: Institute for Healthcare Improvement.
[7] Microsoft Global Evidence Management System (2004) Healthcare Clinic Saves Money and Improves Quality of Care with Tablet PC Solution. Customer Solution Case Study. Redmond, Wash.: Microsoft Corporation.
In response to this concern, Congress in 2000 directed the federal government to conduct a three-year Physician Group Practice (PGP) Demonstration to test how physician payment incentives might promote greater coordination of care, more efficient service delivery, and improved health outcomes for Medicare beneficiaries.
The Centers for Medicare and Medicaid Services (CMS) selected 10 large physician groups in 2005 to participate in Medicare's first pay-for-performance initiative for physicians. This case study examines the early experience of one participating site, the Marshfield Clinic.
Objective: The Marshfield Clinic's mission is "to serve patients through accessible, high-quality health care, research, and education." Strategically, the clinic seeks to achieve the Institute of Medicine's six aims for a high-performance health system: safe, timely, effective, patient-centered, efficient, and equitable health care. The clinic joined the PGP Demonstration to learn how performance incentives could advance these goals, in particular by leveraging information systems to support a population health approach to patient care.
Organization and Project Leadership: The Marshfield Clinic, founded in 1916, is a multi-specialty group practice. Its 730 physicians and 6,000 staff serve more than 360,000 patients at 41 ambulatory care sites located in 35 Wisconsin communities. The physician-governed, nonprofit organization serves all who seek care, regardless of their ability to pay.
Theodore Praxel, M.D., M.M.M., Marshfield Clinic's medical director for quality improvement and care management, and Marilyn Follen, R.N., M.S.N., administrator of quality improvement and care management, oversee the clinic's participation in the PGP Demonstration.
Implementation Timeline: The Medicare PGP Demonstration began on April 1, 2005, and will run through March 31, 2008.
The Marshfield Clinic has electronically coded clinical information on all patients dating back to 1960 and has operated an electronic medical record (EMR) of growing sophistication since 1985. The EMR captures patients' diagnoses, procedures, medications, test results, radiology images, and physicians' notes. Clinic physicians began using wireless, tablet-style personal computers for electronic prescribing and dictation in 2003. The clinic plans to completely eliminate the use of paper charts (a process already underway) by the end of 2007, when every physician is scheduled to have a tablet PC.
Target Population: Elderly and disabled fee-for-service Medicare beneficiaries are assigned to the demonstration if they receive the plurality of their outpatient care from a participating physician group.
Key Measures: The PGP demonstration uses 32 performance measures, a subset of those developed by the American Medical Association's Physician Consortium for Performance Improvement for the Medicare Doctors' Office Quality project being conducted by Medicare's Quality Improvement Organizations. Measures are added incrementally over the three-year demonstration, as shown in the Table below.
| Year 1 Measures | Year 2 Measures (additional to Year 1) |
Year 3 Measures (additional to Year 2) | |
| Diabetes | Congestive Heart Failure | Coronary Artery Disease | Hypertension and Preventive Care |
| HbA1c management HbA1c level Blood pressure management Lipid measurement LDL-cholesterol level Urine protein testing Eye exam Foot exam Influenza vaccination Pneumococcal vaccination |
Left ventricular function assessment Left ventricular ejection fraction testing Weight measurement Blood pressure screening Patient education Beta-blocker therapy ACE Inhibitor therapy Warfarin therapy Influenza vaccination Pneumococcal vaccination |
Antiplatelet therapy Drug therapy for lowering LDL cholesterol Beta-blocker therapy (for prior heart attack) Blood pressure measurement Lipid profile LDL-cholesterol level ACE Inhibitor therapy |
Blood pressure screening Blood pressure control Plan of care Breast cancer screening Colorectal cancer screening |
| HbA1c = Hemoglobin A1c; LDL = low-density lipoprotein; ACE = angiotensin-converting enzyme |
A participating physician group may earn a bonus of up to 80 percent of any Medicare cost-savings that it achieves that exceed 2 percent of its expenditure target (the group is not penalized if it does not meet its target). The expenditure target is based on the practice's own base-year costs inflated by the risk-adjusted annual expenditure growth rate for a comparison group of Medicare beneficiaries. If the PGP qualifies for a bonus, a portion (30 percent the first year and rising to 50 percent by the third year) is tied to the physician group's performance on quality targets. Medicare retains the remaining 20 percent of savings achieved by the PGP plus any bonus set aside for quality performance that is not earned by the PGP. [2]
Process of Change: Marshfield Clinic is using electronic tools to improve workflow and promote quality improvement in several ways. Ongoing information technology investments, including the cost of its electronic medical record (EMR), comprise about 3.5 percent of its annual revenue.[3]
- The EMR generates an Intervention List (i-List) for each physician that identifies high-risk patients with multiple chronic conditions. This integrated approach is superior to separate disease-specific registries as it helps physicians proactively plan visits and follow-up for patients needing a host of evaluation and monitoring services. (In the future, the system also will enable physicians to send patient reminders.)
- At patient visits, medical assistants use tablet PCs to enter symptoms and vital statistics directly into the electronic record, which physicians then review. Laboratory test results, X-rays, and other imaging studies are available electronically for physician or consulting specialist review, thus eliminating delays for document or film transfers and the duplicate testing that often results from missing information.
- An electronic "dashboard" organizes critical information including diagnoses, vital statistics, current medications, drug allergies and reactions, appointments, and reminders for preventive and chronic care services (an application known as PreServ). Physicians can accelerate the monitoring schedule as appropriate, including cases where the patient needs closer follow-up after abnormal test results.
- Electronic prescribing by physicians with tablet PCs (which includes handwriting recognition software) and computer-printed prescriptions reduce problems related to illegible handwriting, thus reducing the potential for medication errors, pharmacy callbacks, and patient time waiting for prescriptions to be filled.
- When a patient calls the 24-hour nurse line, nurses refer to their electronic record and tailor advice to their care plan. Nurses perform triage using physician-approved online guidelines and, when appropriate, make appointments for the patient to see their physician the next day (at selected clinics). The call is documented in the electronic record and the patient's physician receives e-mail notification for reference.
The Marshfield Clinic is redesigning its appointment scheduling process to increase the accessibility and continuity of primary care. [4] By reserving some appointments for daily needs, patients are more likely to see their usual physician in a timely manner when they need care. This approach may help prevent complications and more costly care for those with chronic conditions such as heart failure. It also eliminates duplicate visits for those who would otherwise see another physician or visit the emergency room and then require follow-up with their regular doctor.
The clinic seeks to involve physicians in quality improvement through several mechanisms, including online guidelines and storyboards, continuing medical education, and coaching by four regional quality-improvement medical directors. The medical directors attend departmental meetings to share performance results and improvement strategies and solicit feedback. Local sites are becoming engaged in redesign efforts that aim to optimize workflows. For example, medical assistants have been trained to conduct diabetic foot exams, with physicians becoming involved only when there is an abnormal result requiring intervention.
Early Results:
- Several clinical performance measures have improved since the clinic implemented EMR reminders and related process improvements. For example, electronically retrievable documentation of diabetic foot exams increased from virtually zero to more than 50 percent of diabetic patients.
- Hospitalizations decreased by 28.7 per 100 person-years among patients enrolled in a pilot of the anticoagulation program, saving $271,014 per 100 patient-years. [5]
- Timeliness of appointment scheduling has improved, as measured by decreased time to the third-next available appointment, an accessibility metric used by many physician groups. [6]
- Satisfaction among patients enrolled in care management programs exceeds 85 percent. Overall patient satisfaction has increased with the implementation of tablet PCs, and anecdotal feedback suggests that patients are generally responding positively to the use of this technology during office visits.
- Through instant access to electronic information and electronic functions such as automated prescription renewals, physicians save an estimated three to seven minutes per patient visit, or 200 to 466 hours in productive time per year. Elimination of paper charts is expected to save an estimated $9 million per year in clerical costs.[7]
A well-developed electronic medical record system can support performance goals by improving physician productivity and patient care, thereby creating clinician demand for EMRs. (Marshfield Clinic clinicians are currently on a waiting list to receive their tablet PCs.) When asked if there is physician resistance to the EMR, Praxel responded, "that is like asking me who [doesn't want to] use electricity or water. It's become an integral part of the practice."
On the other hand, Praxel cautions that EMR adoption should not viewed as a solution in and of itself. "It's a good tool; but one has to take the data and turn it into actionable information." In particular, the clinic is seeking to leverage IT to promote population health through the feedback of performance data to physicians that enables them "to examine [their] practice and take action to improve," says Follen.
"Physicians in general are used to thinking about patients one at a time," Praxel explains. "While we don't ever want to lose the individual aspect of patient care, we're trying to close the loop from the individual back to the population to raise awareness of how the physician is doing with a given disease or constellation of diseases for their entire panel of patients." Physicians can then apply that broader perspective to improve the care of individual patients.
Despite the emphasis on information technology, Praxel sees the clinic's quality improvement process as primarily non-technical in nature. "We're using IT to help create transformational change in how we deal with patients," he says. "That's why we have regional quality improvement directors meeting with departments, explaining why it's necessary to alter the way care is currently delivered to patients, and getting physicians involved in change as it's occurring," he says.
Follen believes that an internally developed care management program fosters physician acceptance of change. "Involve your providers so it is truly an extension of their practice and so patients don't feel that they are being dismissed by the physician and managed by a nurse," she says. By integrating care management into clinic operations, "everyone feels it is a team effort to improve care for the patient and the population."
Moreover, process improvements must be tailored to the local environment rather than a forcing a "one-size-fits-all" approach. For example, what is optimal in a large clinical center may be very different from what is optimal in a small one, Praxel points out.
Implications: Although an internal systems development process was well suited to Marshfield Clinic's circumstances, it may not be the appropriate path for the smaller physician groups in which most physicians practice and most patients receive their care. Praxel, who spent 10 years working in a group of 10 to 12 providers before joining the Marshfield Clinic, isn't confident that many small groups have the time and resources necessary to duplicate the clinic's homegrown infrastructure. Other strategies to adapt and embed off-the-shelf systems might be an option for smaller practices.
The capitation-like incentive structure created by the PGP Demonstration may not be broadly applicable to physician practice outside large groups. For example, the Marshfield Clinic does not own or control hospital facilities and faces barriers in proactively coordinating inpatient care because of legal concerns about sharing patient information between separate entities. This may put the clinic at a disadvantage in controlling global costs in comparison to fully integrated PGPs (that include both physician groups and hospitals) participating in the demonstration. It also highlights the potential limits of pay-for-performance in achieving transformational system change.
Establishing common definitions will be critical to furthering the adoption of information technology and performance measurement. The clinic, says Praxel, found "different definitions for monitoring quality from different payers to be taxing and costly." The Institute of Medicine recently recommended that the federal government establish a standards body to address this concern. But, until electronic systems are interoperable, there will continue to be wasteful duplication and gaps in information as people move from one system to another.
For Further Information: Contact Theodore Praxel at [email protected] or Marilyn Follen at [email protected].
References
[1] L. Casalino et al. (2003) External Incentives, Information Technology, and Organized Processes to Improve Health Care Quality for Patients with Chronic Diseases. Journal of the American Medical Association 289, 434–441; A. J. Audet et al. (2005) Physicians' Views on Quality of Care: Findings from the Commonwealth Fund National Survey of Physicians and Quality of Care. New York: The Commonwealth Fund.
[2] RTI International (2005) Physician Group Practice Demonstration Quality Measurement and Reporting Specifications, Version 2. Baltimore: Centers for Medicare & Medicaid Services.
[3] Marshfield Clinic (2003) Integrated Computer Technologies at Heart of Ability to Care for Patients Effectively. Cattails (Nov/Dec).
[4] M. Murray and D. Berwick (2003) Advanced Access: Reducing Waiting and Delays in Primary Care. Journal of the American Medical Association 289, 1035–1040.
[5] M. Hillman (2002) Testimony Before the Subcommittee on Health of the House Committee on Ways and Means, Hearing on Promoting Disease Management in Medicare (April 16). Washington, D.C.: U.S. House of Representatives.
[6] Marshfield Clinic—Indianhead Center (2006) Improving Access in Primary Care—Virtually. Improvement Report. Boston, Mass.: Institute for Healthcare Improvement.
[7] Microsoft Global Evidence Management System (2004) Healthcare Clinic Saves Money and Improves Quality of Care with Tablet PC Solution. Customer Solution Case Study. Redmond, Wash.: Microsoft Corporation.