How do we characterize the current era in US health care? Perhaps “a time of increasing costs and decreasing health insurance”. Equally accurate would be “the age of specialism”. Hospitals have reorganized into specialty service lines;1 profitable specialist-owned surgicenters, endoscopy units, and imaging centers have multiplied;2 market-dominant single-specialty groups can negotiate high fees from commercial insurers.3 The gap between the income of many specialists and that of primary care physicians is wide and growing wider, leading a growing number of US medical school graduates to avoid primary care careers.4

Who is responsible? In most developed nations, the government has a policy that supports primary care as a vibrant foundation for their health care system. The United States, in contrast, has no primary care policy and for decades has relegated physician compensation and workforce planning to the private market and the (specialist-rich) medical profession and academic medical community.

In this issue, Lasser et al. demonstrate that governmental payments are, in large part, responsible for the specialist–generalist income gap.5 Moreover, Medicare’s physician payment structure forms the basis for private insurer fees. If government payment structures, related to the lack of a governmental primary care policy, are responsible for the growing crisis in primary care, it follows that the government—in particular the federal government—is capable of fixing the problem.

How might the federal government invest in a robust primary care sector? Six proposals come to mind.

  1. 1.

    Reforming the CMS-RUC nexus

    Fee-for-service Medicare assigns each physician-provided service a relative value unit (RVU) and multiplies the RVU by a monetary conversion factor (CF) to derive the approved fee. For example, in 2007 a 99214 complex primary care visit had a RVU of 2.58, while a colonoscopy enjoyed a RVU of 5.84. Multiplying by the 2007 Medicare CF of 37.9, the primary care visit garners an average Medicare fee of $97.78 while the average colonoscopy is worth $221.34; both services take about 30 minutes of physician time. In the early 1990s, RVU values were set by the Resource Based Relative Value Scale system (RBRVS) and are updated every 5 years by a committee of the American Medical Association (AMA) and specialty societies—the Relative Value Scale Update Committee (RUC)—which recommends RVU revisions to the Centers for Medicare and Medicaid Services (CMS). While primary care physicians provide about half of Medicare physician visits, they compose 15% of the RUC’s voting members.4 The RUC and CMS perpetuate the inequity between undervalued cognitive and overvalued procedural/imaging payment.6

    The Medicare Payment Advisory Commission (MedPAC), an independent agency that advises Congress on Medicare, asserts that CMS relies too heavily on the physician specialty societies which participate in the RUC.6 In its March 2008 report, MedPAC recommended that CMS establish a group of experts separate from the RUC—including members who do not personally benefit from Medicare payment rates—to identify overvalued services.6

    The Oregon Medical Association (OMA) recently passed a resolution calling for 1) the RUC to make primary care voting representation at least equal to the proportion of primary care physicians in the workforce and 2) the AMA to work with the RUC to address fair primary care compensation. In 2007, the AMA House of Delegates referred OMA’s resolution to the AMA Board of Trustees, who subsequently authored a report supporting the work of the RUC and maintaining that “the RUC was designed to be an expert panel, rather than a representative committee.”7 In 2007 the RUC reviewed a proposal for one new primary care seat but concluded that the RUC’s current expertise was sufficient to competently review RVU values.

    Over the years, CMS has accepted almost all RUC recommendations; thus, it is as responsible—if not more so—for the gaping imbalance between cognitive and procedural/imaging service values. Congress or CMS could address the dominance of the RUC on physician payment. Policy reformers concerned with the impact of the primary care-specialty income gap on the threatened primary care physician workforce might do better to focus their energies on Congress and CMS rather than to work through the AMA.

  2. 2.

    Splitting the SGR

    In 1997, Congress created the sustainable growth rate (SGR) formula to control Medicare spending by setting yearly targets for total Medicare physician expenditures.8 Each year, if total physician expenditures exceed the target for that year, the SGR mandates Congress to reduce the conversion factor (CF) in order to bring Medicare physician spending back into line. Pressured by the physician lobby, Congress has repeatedly overridden the SGR targets, preventing reductions of the CF and allowing rapid increases in physician expenditures. Advocates of fiscal restraint counter that stringent enforcement of the SGR is needed to keep Medicare solvent. Lost in this debate is an appreciation of how the SGR approach exacerbates the gap between primary care and specialist earnings.

    How does SGR penalize primary care? Physician services can be divided into two groups: evaluation and management (E&M) services (most commonly primary care office visits) and non-E&M services, including surgical, diagnostic (e.g., colonoscopy), and imaging (e.g., CT scans) procedures mostly billed by specialists. The SGR applies a single CF to all physician services. But the volume of non-E&M services has grown far more rapidly that the volume of E&M services, enriching specialists and, in 2006, accounting for 86% of physician payment overruns above the SGR target.9 The number of office visits for established Medicare patients increased by 12% from 2000 to 2005, while the number of Medicare-billed colonoscopies increased by 40%, and CT scans by 65%.4 With SGR using one CF for all services, primary care physicians are penalized when large increases in spending for specialized services drive down—or keep flat—the CF applied to E&M and non-E&M services alike.

    What if Congress created two SGR pools, one for E&M and the other for non-E&M services? In this scenario, the CF for each pool would rise or fall based on expenditure trends within that pool. The split SGR would hold procedural and imaging specialists responsible for the volume growth of their services; the non-E&M CF would go down if the non-E&M pool overspends, while the E&M CF would go up if cognitive services continue slow volume growth. This proposal not only helps primary care reimbursement; it also puts a brake on Medicare physician expenditures by delivering consequences (a lower CF) for specialist procedural and imaging overspending. A detailed proposal for a split SGR was published in a recent Health Affairs blog: "Splitting Medicare’s Sustainable Growth Rate: A Proposal to Strengthen Medicare and Primary Care."9

  3. 3.

    Upward adjustment of primary care fees

    MedPAC has floated the idea of maintaining the RBRVS and SGR systems intact, but making an upward adjustment of fees for cognitive primary care services. This would tilt Medicare payment toward primary care and (because the SGR is a fixed pie—if the primary care slice increases, the specialist portion shrinks) away from specialist-provided procedural and imaging services.10

  4. 4.

    New primary care payments

    Primary care payment could be increased through a blend of visit-based fees and other reimbursement modes, for example, care coordination payments for patients with complex healthcare needs. MedPAC recommended that a care coordination payment be added to the physician fee schedule to compensate physicians for the between-visit time spent on coordinating care.11 The Geriatric Assessment and Chronic Care Coordination Act of 2007, supported by over 30 professional and consumer organizations, has also proposed a care coordination payment for Medicare patients.

  5. 5.

    Elimination of fee-for-service primary care payment

    RBRVS and SGR reform assume the continuation of fee-for-service payment for primary care. A more profound reform involves the elimination of visit-based fee-for-service primary care payment and its replacement by a comprehensive per patient payment that is risk adjusted and combined with substantial performance-based rewards (using patient experience and clinical measures) to encourage quality and access. This comprehensive payment for comprehensive care proposal would pay considerably more than the capitation payments common in the 1990s.12 The proposal argues that primary care payment should not be visit-based because care coordination takes place between visits, and phone/e-mail/Web encounters should be provided as alternatives to face-to-face visits. Commercial insurers are being approached to pilot-test this payment reform.

  6. 6.

    The “patient-centered medical home”

    According to this popular concept, primary care practices that can demonstrate certain features (e.g., electronic medical records, a registry to track clinical measures, coordination of care, prompt access to care, and other improvements) would be designated as medical homes and rewarded with increased reimbursement. The National Committee on Quality Assurance has created criteria for medical home designation. In 2006, Congress authorized Medicare medical home demonstrations, requiring Medicare to organize medical home projects in up to eight states. Thirty state Medicaid programs have expressed interest in promoting medical homes. Commercial health plans in several states have joined forces to pilot new payment modes for medical home-designated practices. Large employers, national health insurance plans, and primary care professional societies have organized a Patient Centered Primary Care Collaborative seeking to strengthen primary care through the medical home concept.13 Precisely how reimbursement might change for a medical home-designated practice will vary depending on the payer (Medicare, Medicaid, or commercial insurer).

    Small primary care practices cannot by themselves accomplish the triple aim of the medical home: enhancing the patient experience of care, improving the health of populations, and reducing per capita costs. Only through structures that integrate small practices with one another and with specialists, hospitals and home care services, can the medical home vision become a reality.14

In conclusion, it appears that the United States has been inching toward a primary care policy over the past 2 years. An impending primary care practitioner shortage is widely recognized. The decreasing ability of patients to find a new primary care practice, or to get a timely primary care appointment, is making headlines. Employers finally understand the virtues of a strong primary care foundation to the health care system. The Government Accountability Office, which advises Congress on policy issues, testified in 2008 that a strong primary care-based system offers higher quality and lower costs, arguing that the current undervaluing of primary care is harmful.15

Talk, however, is cheap. Since, as Lasser et al. demonstrate, the government is largely responsible for primary care’s doldrums, then the government should step up to the plate and do something about it. The six concrete proposals discussed here would be steps in the right direction. Ultimately, the United States needs to join other developed nations in proclaiming a primary care policy to the effect that “The federal government will take all necessary steps—including payment reform and physician workforce planning—to ensure that a robust primary care sector thrives as the foundation of US health care.”