The perverse incentives of fee-for-service reimbursement have been discussed at length during the health care reform debate. The Institute of Medicine has estimated that 30 to 40 cents of every health care dollar is spent on inappropriate, duplicative, or ineffective care, costing the nation between $600 and $700 billion annually.1 It is difficult to see how health care costs can be contained without first reforming how the United States pays for health care. In fact, health care reform that improves access to care without tackling reimbursement reform will only exacerbate the health cost crisis.
In the current economic recession, academic medicine faces serious financial challenges, but it is the threat to the clinical revenues, which make up the largest component of an academic health center's (AHC's) budget, that carries the most concern. If, as a result of health reform, clinical reimbursement declines significantly, academic medicine's important missions of caring for the nation's most vulnerable patients, educating the next generation of health care professionals, and conducting research to improve the lives of patients may be endangered. Because of its central importance to health care reform, payment reform is an issue that AHCs cannot afford to ignore. This commentary will explore the potential impact on academic medicine of the major payment reform proposals under discussion by Congress and provide a series of recommendations to help AHCs prepare for the challenges and opportunities of payment reform.
The Three Prominent Proposals for Payment Reform
Proposals to move beyond the fee-for-service reimbursement model can be divided into three broad approaches. The first plan involves payment of a prenegotiated fee to a provider to coordinate the health care needed by a patient during a defined period of time. The second proposal involves payment based on the quality of patient outcomes achieved. A third approach revolves around paying for discrete episodes or bundles of care—for example, the care surrounding a surgical procedure like a hip replacement. These three approaches are not mutually exclusive and, in fact, can be used in combination or combined with fee-for-service reimbursement to create a coherent system of reimbursement that establishes effective incentives for providers and patients.
Care coordination and the advanced medical home
The advanced medical home (AMH) concept originated in the 1960s but has only recently attracted national attention. The AMH model seeks to encourage a greater role for primary care, based on research showing that patients with an established primary care provider have better health outcomes,2 and to address the shortage of primary care providers by enhancing reimbursement for primary care services. In the AMH proposal, a primary care practice accepts long-term responsibility for the health care needs of a patient population, serving as the first point of contact for new health problems, providing comprehensive care for issues not requiring specialists, and coordinating care requiring referral outside the practice. In exchange, the practice receives a global per-member-per-month capitation payment or is paid fee-for-service for the discrete services provided plus a supplemental care coordination payment. Practices may also receive bonus payments for achieving quality standards. Care is typically provided by a team, making extensive use of nonphysician providers. An electronic health record is critical for coordinating care, providing decision support, and tracking quality metrics. The number of patients cared for by the practice is kept small enough to give physicians more time with patients, provide recommended preventive care services, and allow patient participation in decisions related to their care, including 24-hour electronic communication. Estimates of the cost-saving potential of the AMH are equivocal; they range from an additional national cost of 0.4% to a savings of 1.2%.3
Demonstrations of the AMH model, sponsored both by the Centers for Medicare and Medicaid Services and by insurers like CIGNA and United Health, are now under way. Some published evidence supports the effectiveness of the AMH proposal. At Group Health Cooperative in Seattle, implementation of the AMH plan reduced emergency room visits by 29% and hospitalizations by 11% while improving quality of care.4 Community Care of North Carolina, a similar project involving networks of community health centers, saved the state's Medicaid program $463M between fiscal year 2003 and 2006.5
Accountability for results
Pay-for-performance is not a new concept, but, although it makes sense to reward providers for desirable outcomes, published results to date have been equivocal.6 However, a new twist on this concept, the accountable care organization (ACO), is gaining traction with lawmakers. An ACO is a tangible or a virtual organization of providers that takes responsibility for the global health care needs of a defined population of 5,000 to 10,000 patients for a specific period of time.7 The organization holds itself to measurable quality standards and, in return, is paid fee-for-service reimbursement for discrete services provided as well as sharing in any savings resulting from improved coordination of care and better patient outcomes.
A variant on the ACO model is the Healthcare Innovation Zone (HIZ) proposed by the Association of American Medical Colleges. Pioneered by Johns Hopkins University in Baltimore, an HIZ is a partnership between an AHC and a group of community primary care providers to leverage the unique research, education, and clinical capabilities of the AHC to care for a population of patients. Demonstration projects relating to the ACO concept have begun in a number of places and are being proposed as part of federal health care reform. As part of health care reform in Massachusetts, for example, a special commission has recommended that fee-for-service reimbursement be replaced by a statewide system of ACOs capable of receiving global payments for the health care services needed by populations of patients.8 To be effective, ACOs require leadership that can functionally integrate disparate groups of providers, manage care efficiently, and make significant investments in IT infrastructure to knit practices together.
In the episode-of-care approach, often called “bundled reimbursement,” all providers caring for a patient during a defined episode of care split a global, prospectively negotiated payment. The payment is expected to cover all services needed by a patient during the episode, including routine and/or recommended care as well as care for any complications that arise. The major emphasis is on reducing fragmentation, avoidable complications, and duplication of services. An example of the bundled reimbursement model is ProvenCare, developed by Pennsylvania's Geisinger Health System. Geisinger charges subscribers a fixed price for coronary artery bypass surgery, which includes the expenses related to the preoperative workup, surgical care, and 90 days of postoperative care. If any complications occur, Geisinger pays the expenses required to address them. ProvenCare has been described as a “warranty” and may represent one of health care's first forays into consumer guarantees common in other commercial businesses. ProvenCare's key components include strong support from senior leadership for innovation in quality and safety, development by the surgeons of a standardized clinical process containing 40 discrete elements of care based on national care guidelines, a system-wide electronic medical record with embedded standardized order sets and clinical decision support that enables physicians to apply a uniform approach to patient care, and a cost-accounting system that accurately reports the costs related to caring for these patients. Outcomes of the ProvenCare project have been excellent: 100% compliance with the care protocol, a 44% decrease in readmissions, and a 16% decrease in length of stay.9
Prometheus, another approach to an episode-of-care payment model, involves the development of evidence-informed case rates for particular conditions based on expert-recommended elements of care.10 Evidence-informed case rates are adjusted for patient comorbidities and also include an allowance for complications. Ten to twenty percent of provider reimbursement is contingent on achieving a set of quality metrics, and another 30% is conditioned on the care rendered to the patient by other providers participating in the case. Thus, the emphasis is on coordination of care, avoiding preventable complications, and achieving quality metrics. The system is being piloted for a number of conditions, including surgical procedures such as hip and knee replacements and episodes of chronic illness such as care of diabetic patients for a one-year period at health systems across the United States.
The Centers for Medicare and Medicaid Services are also conducting the Acute Care Episode, a three-year demonstration in five U.S. cities, in which Medicare pays participating hospitals a discounted, prenegotiated, fixed price for elective orthopedic and cardiac surgery procedures. Patients are paid incentives ranging from $250 to $1,157 for receiving care at participating hospitals.
A concern with bundled payment is that it does not attempt to control the volume of services provided. Although opinion on the cost-saving potential of bundled reimbursement is not uniform, many think that it can reduce spending; by one estimate, it would result in a 5.4% reduction in national health care spending.3
Disadvantages and Advantages of the Three Proposals for Payment Reform
Given the complexity and financial importance of reimbursement reform, the passage of health care reform legislation is likely to direct the creation of pilot programs involving bundled payments, the AMH model, and ACOs to gain experience in what works before large-scale adoption of new models is mandated. However, long-term payment reform seems inevitable, and given the importance of clinical revenue to their survival, prudent AHCs should begin preparing to function in a new reimbursement paradigm.
It will not be easy for academic medicine to deal with innovative payment methodologies. First, most AHCs are large, complex, and fragmented organizations with a significant number of autonomous operating units. Second, the academic environment prizes individual and intragroup autonomy. Collaboration across institutions is not second nature. Third, although AHCs have often been pioneers in the adoption of electronic medical records, many still lack the robust information technology infrastructure, including cost-accounting systems, that is required for payment reform. Fourth, medical costs in academic centers are frequently higher than in community hospitals that do not have the added responsibilities of health care education, research, and quaternary clinical care, a potential negotiating disadvantage with third-party payers. Finally, many academic centers are not customer service organizations that give first priority to patient care over the more academic pursuits of research and education.
Still, AHCs have many advantages that could position them to be successful in the new reimbursement paradigms. First, most academic centers are composed of large, multispecialty physician groups and affiliated hospitals and health systems. If senior management can find ways of functionally integrating these assets into a coherent system that can effectively coordinate the care of patients, the advantage of size and scope would be significant. Second, academic centers have a strong tradition of research and, as a result, are ideally situated to study the outcomes of innovative care delivery models. Third, academic centers frequently have a reputation for high-quality, cutting-edge patient care that is attractive to patients and payers alike. Many AHCs are the only providers of highly specialized services in their markets, which increases their negotiating power with payers. Finally, AHCs are often leaders in their communities and regions and, as such, can help to organize coalitions to experiment with new payment methodologies.
Recommendations for AHCs
To succeed in a new system of clinical reimbursement, AHCs will need
- senior leadership that creates and visibly supports a culture of quality, innovation, and patient safety,
- physician leadership in and commitment to process improvement and care redesign efforts,
- a system-wide electronic health record that allows the coordination of care across the organization and an information technology staff that can build customized care protocols,
- a talented staff that understands quality improvement and is available to help providers implement process-of-care improvements and design and track outcomes metrics,
- a cost-accounting system that can accurately assign costs to the care of individual patients,
- collaborative relationships between physician groups and hospitals such that patient care can be coordinated seamlessly across systems and financial issues can be negotiated fairly and expeditiously,
- a patient-centered approach to care delivery that attracts patients to academic centers for their health care needs,
- relationships with third-party payers that allow the negotiation of novel payment arrangements, and
- educational programs that emphasize patient safety and quality as important components of the knowledge base of physicians and other health providers.
Nearly everyone agrees that the days of pure fee-for-service reimbursement are numbered. Given, then, that payment reform is inevitable, academic medicine must lay the foundation for these new payment models that will become the norm in the not-too-distant future. AHCs should use their prominent position in their regions and nationally to lead the way forward and help to shape how new payment models are rolled out across the country.
1Reid PP, Compton WD, Grossman JH, Fanjiang G, eds; National Academy of Engineering and Institute of Medicine of the National Academies. Building a Better Delivery System: A New Engineering/Health Care Partnership. Washington, DC: National Academies Press; 2005. Available at: http://books.nap.edu/openbook.php?record_id=11378&page=1
. Accessed December 3, 2009.
2Starfield B, Shi L, Macinko J. The contribution of primary care to health systems and health. Milbank Q. 2005;83:457–502.
3Hussey PS, Eibner C, Ridgley MS, McGlynn EA. Controlling U.S. health care spending—Separating promising from unpromising approaches. N Engl J Med. 2009;361:2109–2111. Available at: http://healthcarereform.nejm.org/?p=2301&querry=home
. Accessed December 3, 2009.
6Rosenthal MB, Frank RG, Li Z, Epstein AM. Early experience with pay-for-performance: From concept to practice. JAMA. 2005;294:1788–1793.
9Casale AS, Paulus RA, Selna MJ, et al. “ProvenCareSM
”: A provider-driven pay-for-performance program for acute episodic cardiac surgical care. Ann Surg. 2007;246:613–623.