Perspective

Academic Medicine: A Key Partner in Strengthening the Primary Care Infrastructure Via Teaching Health Centers Richard E. Rieselbach, MD, Byron J. Crouse, MD, Katherine Neuhausen, MD, MPH, Thomas J. Nasca, MD, and John G. Frohna, MD, MPH

Abstract In the United States, a worsening shortage of primary care physicians, along with structural deficiencies in their training, threaten the primary care system that is essential to ensuring access to high-quality, cost-effective health care. Community health centers (CHCs) are an underused resource that could facilitate rapid expansion of the primary care workforce and simultaneously prepare trainees for 21stcentury practice. The Teaching Health Center Graduate Medical Education (THCGME) program, currently funded by the Affordable Care Act, uses CHCs as training sites for primary-care-focused graduate medical education (GME).

A

robust primary care infrastructure is essential to provide high-quality, cost-effective health care.1 Strengthening the faltering primary care infrastructure of the United States is crucial to enhancing access to care, eliminating health disparities, reducing health care costs, and addressing the federal budgetary crisis. The most formidable obstacle to restoring the U.S. primary care infrastructure is the critical and worsening shortage of primary care physicians, especially those willing to work with patients who are underserved and often live in low-income rural and urban communities. The shortage of primary care providers is already imposing serious limits on access to care,2 and to make matters worse, the primary care physicians graduating from residency Please see the end of this article for information about the authors. Correspondence should be addressed to Dr. Rieselbach, Department of Internal Medicine, University of Wisconsin School of Medicine and Public Health, Health Sciences Learning Center, 750 Highland Ave., Madison, WI 53705; telephone: (608) 263-2552; e-mail: [email protected]. Acad Med. 2013;88:1835–1843. First published online October 14, 2013 doi: 10.1097/ACM.0000000000000035

The authors propose that the goals of the THCGME program could be amplified by fostering partnerships between CHCs and teaching hospitals (academic medical centers [AMCs]). AMCs would encourage their primary care residency programs to expand by establishing teaching health center (THC) tracks. Modifications to the current THCGME model, facilitated by formal CHC and academic medicine partnerships (CHAMPs), would address the primary care physician shortage, produce physicians prepared for 21st-century practice, expose trainees to interprofessional education in a multidisciplinary environment, and

facilitate the rapid expansion of CHC capacity.

programs today are not well prepared for modern clinical practice.3 This poor preparation is not due to a lack of talent on the part of trainees or a lack of effort on the part of their teachers; rather, it is due to the structural deficiencies of a predominantly hospital-based approach to primary care training.3

than merely maintaining the status quo and hoping for a better result.

While graduate medical education (GME) primary care training programs struggle to expand to meet the burgeoning need for primary care physicians in the United States, community health centers (CHCs) are growing at a rapid rate. Currently, over 1,200 CHCs, dispersed throughout all 50 states, provide care for more than 20 million patients.4 By 2019, these centers could reach over 40 million underserved children and adults, many of whom will be newly insured through the Affordable Care Act (ACA).5,6 Further expansion of these centers, a vital element of the primary care infrastructure, is severely constrained not only by the currently insufficient primary care workforce2 but also by the enormous difficulties many CHC physicians face in obtaining subspecialty consultation.7 The United States needs a more creative approach

Academic Medicine, Vol. 88, No. 12 / December 2013

To succeed, CHAMP THCs require a comprehensive consortium agreement designed to ensure equity between the community and academic partners; conforming with this agreement will provide the high-quality GME necessary to ensure residency accreditation. CHAMP THCs also require a federal mechanism to ensure stable, long-term funding. CHAMP THCs would develop in select CHCs that desire a partnership with AMCs and have capacity for providing a community-based setting for both GME and health services research.

Collaboration between the academic medicine community and CHCs holds the key. We believe that teaching hospitals (academic medical centers [AMCs]) should partner with select CHCs to develop a unique type of teaching health center (THC). The resulting model could have major long-term benefits for patients and communities while simultaneously facilitating immediate expansion of CHCs, thereby contributing to medical cost control. We envision the “CHC and academic medicine partnership” or CHAMP developing with CHCs that desire integration with AMCs to access subspecialty care,7 that have capacity for providing a communitybased setting for both GME and health services research, and that exhibit other assets required for a successful partnership.8 THCs are high-value settings for primary care training and workforce expansion.2 By providing mission-driven medical school graduates with training in the care of vulnerable patients alongside the guidance of supportive mentors, THCs

1835

Perspective

are essential to addressing the primary care workforce shortage in underserved communities.2 To illustrate, research has shown that family physicians who trained at CHCs were 2.7 times more likely to work in underserved settings than family physicians trained in other settings.9 THCs prepare residents to work in underserved settings by providing education in community medicine, population health, cultural competency, mental health, and geriatrics.2 In addition to preparing graduates to care for medically and socially complex populations, THCs train primary care physicians in innovative delivery models including patient-centered medical homes (PCMHs) and interdisciplinary teams.10 We believe that the collaboration represented by CHAMPs would strengthen the U.S. primary care infrastructure by establishing patient care, teaching, and health services research programs within local communities,11 thereby advancing an important societal mission.12 Because of their strong, established community relationships, CHCs are ideal partners to collaborate with AMCs in widening the primary care pipeline as an essential step to building a more robust primary care infrastructure. In addition to providing additional resources for CHCs and the patients they serve, CHAMPs could link the academic resources of AMCs with the community expertise of CHCs to develop programs that prepare students from underserved communities to enter and succeed in medical school. Thus, CHAMPs would help develop the workforce serving underserved communities for not only the short-term future but also in the long-term. The ACA provides funding for a limited number of THCs via a well-designed and extensively evaluated program designated the Teaching Health Center Graduate Medical Education (THCGME) program.10 However, this program does not fully meet the rapidly evolving educational and health care delivery needs of the United States. Building on the success of the THCGME, a new multiprofessional, expanded THC model is needed to create the innovations necessary to expand both primary care physician education and CHC capacity for clinical service delivery in participating CHCs. Both are required

1836

elements for successful implementation of the ACA. In complementing the existing THCGME program, the CHAMP THC model we propose offers seven innovative features: 1. concurrent training of residents in three primary care disciplines (family medicine, pediatrics, and internal medicine) in order to provide multiprofessional training and expand access, 2. a specific role for each primary care specialty, 3. a unique educational environment that will rejuvenate interest in primary care, 4. the potential for the development of CHAMP Medicaid accountable care organizations (ACOs) that effectively integrate CHCs with subspecialty care, 5. a well-conceived and executed CHAMP consortium agreement designed to preserve the culture and financial integrity of CHCs while also achieving effective governance and providing the high-quality GME necessary to ensure residency accreditation, 6. the provision of incremental residency positions for both the AMC and the CHC, along with a funding mechanism that assures the CHC equal financial leverage in the consortium with the AMC, and 7. long-term funding via a stable and sustainable federal mechanism. The CHAMP THC program that we propose will help the United States build a more effective primary care infrastructure and, in turn, prepare for the coming implementation of the ACA. Here, we describe each of the seven unique features of our proposed model.

capacity of the participating CHCs,2 thereby expanding access to and reducing the cost of health care to the community. In addition, we anticipate that many of the CHAMP THC graduates (in all three primary care specialties), encouraged by the National Health Service Corps (NHSC) debt repayment incentive, would subsequently opt to join a CHC as clinical staff.2 Expanded CHC capacity through increased medical staff is critical. Without creative initiatives, it is difficult to see how CHCs will close the gap between demand and supply of primary care services. A 2006 survey of 890 CHCs revealed that these clinics needed at least 400 more physicians and that 40% of vacancies remained open longer than seven months.13 Expansion of a multispecialty pool of primary care physicians is crucial especially if CHCs hope to serve the greatly expanded number of patients after implementation of the ACA.5 The success of the ACA (which makes substantial investments in the expansion of CHCs) depends on the CHCs that will serve as major sources of primary care for the millions of Americans living in underserved areas who will gain coverage under the act.5,6 A recently published report by the Kaiser Commission on Medicaid and the Uninsured emphasized the critical role played by CHCs in Massachusetts health care reform.13 CHCs should be able to help ensure that the ACA’s insurance expansion is cost-effective because these practice sites have sought to deliver low-cost, community-oriented primary care since their inception. The CHAMP THC model provides an opportunity for CHCs to enhance their cost-saving capacity, as described in List 1. The Role of Each Primary Care Discipline in a CHAMP THC

The Value of Concurrently Training Residents of Three Primary Care Specialties in a Single THC Setting

Involving residents from three primary care specialties—family medicine, pediatrics, and general internal medicine—in CHAMP THCs would enhance primary care training by engendering interprofessional respect and collaboration. Also, concurrent training would immediately increase the clinical

In the future, primary care physicians will increasingly lead teams that constitute a PCMH. The capability to serve as a PCMH will be an important criterion for CHCs selected to serve as a CHAMP THC.2 Most PCMHs employ a team approach to care that involves primary care physicians, nurse practitioners, physician assistants, pharmacists, care coordinators, dentists, social workers, psychologists, dieticians, and other health care professionals. This

Academic Medicine, Vol. 88, No. 12 / December 2013

Perspective

List 1 Enhancement by Community Health Center and Academic Medicine Partnership (CHAMP) Teaching Health Center (THC) of Community Health Center (CHC) Cost Savings • Use of senior residents and their supervisory faculty as primary care providers for patients, funded by Medicare graduate medical education support • Decreased hospital admissions and emergency department visits for conditions that can be treated in ambulatory care settings resulting from improved continuity, comprehensiveness, and extended hours • Decreased hospital readmissions resulting from coordinated care and improved care transitions • Decreased laboratory and imaging costs resulting from mandated teaching of appropriate resource utilization • Decreased fragmentation of care through effective access to and utilization of consultations • Malpractice liability protection for CHC physicians under the Federal Tort Claims Act • Reduced CHC expenditures on pharmaceuticals through the Federal Drug Pricing Program • Effective chronic disease management and prevention through patient-centered medical home (PCMH) practices • Provision of integrated dental and behavioral health services • Advanced electronic medical record and PCMH development, funded by recent federal programs • Elimination of billing costs, administrative cost for managed care, or both through global payment if the CHAMP THC establishes an accountable care organization (ACO)26 • Global payment and ACO incentives promoting cost-effective care26

team-based approach likely represents the practice environment of the future. A THC is an ideal setting for implementing the PCMH approach and for developing the core competencies required to master interdisciplinary collaboration. As key members of a PCMH team, primary care resident physicians will serve as a personal physician, especially for very complex patients, serve as a consultant and educator for other team members, and play a key role in data analysis for quality improvement and practice redesign. Primary care physician trainees in CHAMP THCs will also facilitate care in the “medical neighborhood” by coordinating the management and care of patients affiliated with the PCMH when interaction with hospitalists, medical or surgical subspecialists, and other health care providers is necessary.7 Each primary care resident will bring valuable expertise to the CHAMP THC team. The family medicine residents will provide care, including preventive care, for families and for individuals across the entire age spectrum. They can also support the needs of other team members by addressing behavioral health issues, as well as women’s health and obstetrical concerns. Pediatric residents will focus on care for infants, children,

and adolescents—a major percentage of Medicaid patients in many CHCs. General internists, per tradition, will care for patients who have complex chronic diseases and/or multiple comorbidities. These physicians will collaborate with family physicians to provide comprehensive care to the most medically and socially complex patients. Residents, working in teams with appropriate faculty supervision, would enable CHCs to offer expanded hours on nights and weekends—much as residents take call in hospitals. This greater access would both provide a valuable, low-cost alternative for patients who turn to emergency departments for the management of many treatable or preventable problems and reduce preventable hospital admissions due to ambulatory-care-sensitive conditions.14 Additionally, stronger therapeutic relationships between patients and physicians developed through continuity in the outpatient setting should also decrease the need for urgent care. An Ideal Educational Environment

Ideally, primary care training in a CHAMP THC should support the

Academic Medicine, Vol. 88, No. 12 / December 2013

development of the competencies and skills required for future practice. The 2011 and 2010 Macy conferences set ambitious goals for reshaping GME to most effectively prepare residents for 21stcentury practice.15,16 List 2 enumerates several ways CHAMP THCs can achieve these goals. The THC teaching structure that we propose CHAMP THCs would follow entails each faculty member supervising four residents in their primary care discipline (as described previously).2 Geriatric medicine To address the “graying” of the United States, all primary care physicians who intend to care for adults should demonstrate competency in caring for older patients. Few young physicians, however, are pursing specialized training in geriatric medicine,17 so most elderly patients will continue to receive their primary, acute, and postacute care from internists, family medicine physicians, or advanced practice registered nurses. Because currently fewer than 7,000 certified geriatricians are practicing, training of other personnel in geriatrics is greatly needed.18 CHAMP THCs, like other THCs, are well positioned to enable geriatric consultants to provide needed training to primary care providers. Nursing home, hospice, and home-based health care should be an important part of a CHAMP THC curriculum. Multispecialty group practice A valuable characteristic of the CHAMP THC educational environment is the opportunity for trainees to master the dynamics of navigating a multispecialty group practice (MSGP). The CHAMP could facilitate subspecialty consultant availability via electronic medical records, telemedicine, and other forms of electronic communications, as well as through the “medical neighborhood” provided by a strengthened relationship with the partnering AMC; thus, in many respects, a CHAMP THC will closely resemble a MSGP.7 Residency training in such a milieu will help primary care residents develop competencies in coordinating subspecialty care in a manner that decreases fragmentation, reduces needless tests and consultations, and optimizes outcomes. By helping residents acquire the communication skills they need to help their patients navigate a complex health care environment, CHAMP THCs should help

1837

Perspective

List 2 Macy Foundation Conference Goals That Community Health Centers and Academic Medicine Partnership Teaching Health Centers Would Address • Providing both interprofessional (family medicine, pediatrics, internal medicine) education and interdisciplinary education (nurses, physician assistants, dentists, pharmacists) for residents • Preparing residents for a specific career by placing them in urban versus rural community health centers (CHCs) • Providing residents with direct faculty supervision and mentoring, formative feedback, and adequate independence • Giving residents the opportunity to provide service to society during their training • Allowing residents to play a role in teaching other trainees, including medical students • Providing residents with training in the patient-centered medical home (PCMH) model • Providing faculty development for medical resident and allied health PCMH trainees • Modeling innovations and reforms in graduate medical education (GME) financing that are based on societal priorities • Enhancing the role of community in GME by consumers’ operational control of CHCs through their greater than 50% membership on CHC boards • Giving residents the opportunity to be involved in nursing homes, hospice, and home care • Providing residents with primary care training in geriatric, women’s, teen, and mental health • Developing residents’ competence in the use of electronic health records (EHRs; CHCs have led implementation of EHRs)

Interest in primary care

• Encouraging residents to select a career choice of high social value through tailoring GME outcomes to social needs

We believe that the attractive educational environment of CHAMP THCs will foster interest in primary care. This is a particularly important priority for internal medicine because, in many programs, less than 20% of graduating residents currently elect to practice primary care. Training within the CHAMP THC track of an existing competitive internal medicine primary care residency program would demonstrate the “academic credibility” of this career path and provide the opportunity for residents to experience a well-structured primary care delivery system. The prospect of debt repayment via the NHSC already associated with THC training is an additional incentive for choosing a primary care career.2,22

• Placing emphasis on cost-effectiveness and high-quality care through providing training in quality parameters • Responding to changes in health care delivery (e.g., accountable care organizations, global payment) • Serving a diverse mix of CHC patients and providing training in cultural competence • Providing residents with training in team leadership, social responsibility, and conflict management • Focusing on population-medicine-centered training, including preventive medicine, community health, and the socioeconomic determinants of health

reduce avoidable hospital readmissions. Effective communication and patient hand-off skills will be especially important in the future, as group practices in the United States are growing at an unprecedented rate of about 10% annually and currently employ close to 140,000 physicians.19 Practice-based quality improvement A typical CHAMP THC curriculum would include a quality improvement and practice redesign course to teach residents how to use data to drive practice improvement and to employ tools such as Plan-Do-Study-Act cycles. Mentored by faculty, CHAMP THC residents could lead interdisciplinary teams of clinic staff in all stages of clinic-based quality improvement initiatives including designing an intervention and collecting and

1838

July 1, 2013.21 The NAS will evaluate resident achievement of milestones at various stages of training. The milestones contain expectations that residents will function as leaders and participate in team-oriented care. In addition to the requisite clinical and professional attributes, the NAS will require residents to demonstrate competence in such areas as cost-effectiveness, health literacy, and the use of health information technology which will help them work to improve the health of individuals and populations. CHAMP THCs are well suited to acquire the infrastructure needed to instill these skills in residents because CHCs require residents to do more for patients and learn how to provide care for a lower-literacy population. Further, CHCs already have nearly universal EMR adoption and already comply with population health measurement and reporting requirements.

analyzing data. Many CHCs already employ experts in communityoriented primary care and quality improvement10; these CHCs would provide rich training environments for residents learning such skills. This combination of didactic teaching and practical experience would prepare residents to participate in and lead data-driven improvement efforts, which are key to high-performing primary care practices.20

NHSC incentives

Competencies New competencies (like those related to quality improvement cycles) will likely be required of all primary care residents to meet the Accreditation Council for Graduate Medical Education (ACGME) milestones of the Next Accreditation System (NAS).21 The ACGME began phasing in the NAS on

The NHSC has implemented a number of changes to its Scholars Program and Loan Repayment Program, which together increase the incentive for training in any THC residency, including a CHAMP THC.22 As the median debt of health professional students continues to mount, the opportunity to receive— above and beyond normal salary and benefits—up to $120,000 for four years of service and then $20,000 a year for each additional year of service to pay off eligible educational debt can eliminate, or at least mitigate, the financial barrier to

Academic Medicine, Vol. 88, No. 12 / December 2013

Perspective

serving disadvantaged and underserved populations as CHC staff. A new NHSC initiative, the Student to Service (S2S) program,23 is an application-based loan repayment program for students in their last year of medical school (when they are more committed to a primary care residency). Accepted students receive $120,000 in four payments beginning in the first year of residency, and in return they are required to provide three years of service at approved NHSC sites. S2S residents initiate loan repayment during residency training, and their service obligation begins after they complete their primary care residency. CHAMP residents who participate in the S2S program could fulfill their service at the THC where they trained, providing continuity of care for their patients. By spending up to half of their time (20 hours per week) teaching, they could fulfill their NHSC obligation24 while helping to meet the faculty needs of the THC. The ACA authorized $1.5 billion across five years to expand the NHSC,25 which is currently offering 100 positions per year in the S2S. Expanding this program to 1,000 positions per year would substantially increase the potential physician workforce for CHCs. The CHAMP Medicaid ACO

The capacity to serve as a Medicaid ACO is another important feature.26 The CHAMP Medicaid ACO would combine the subspecialist expertise, state-of-theart medical technology, and inpatient care of AMCs with the primary care expertise of CHCs. The subspecialty component of this MSGP would be facilitated by electronic consultation and point-of-care decision support tools.26 This computerenabled multispecialty environment would be provided and financed by the ACO in conjunction with the AMC group practice, using global payments without the excessive profit margins associated with Medicaid managed care plans.26 Training in CHAMP ACOs that engage in clinical innovations to decrease costs and increase quality would prepare THC residents to practice in the delivery systems of the 21st century. Residents would learn how to factor cost and quality into their clinical decisions. They could participate in innovative care coordination and disease management programs and (as mentioned) learn

quality improvement skills and use point-of-care decision support tools. Developing these competencies would prepare CHAMP THC residents to succeed under the value-based payment models, such as pay-for-performance, bundled payments, and capitated payments, that will be increasingly common in their future practice. A CHAMP Medicaid ACO may need to forge partnerships with nursing homes, behavioral health providers, and social service agencies to meet the complex needs of vulnerable populations. These partners would provide valuable training experiences for CHAMP ACO primary care residents. Further, because untreated behavioral health conditions and social factors are often major drivers of the preventable hospitalizations and readmissions, such strategic partnerships would help the CHAMP Medicaid ACO achieve cost savings.27 The shared savings generated by the CHAMP ACO through the cost savings achieved by decreasing preventable admissions would also benefit the CHC. Reinvestment of shared savings would provide an important revenue stream to support innovation by both the AMC and CHC. Adopting the ACO payment model increases the financial viability of CHAMP THCs by reducing the financial barriers to their implementation. CHAMP Medicaid ACOs are ideally positioned to take advantage of Medicaid Integrated Care Models, developed by the Centers for Medicare and Medicaid Services to allow states to develop Medicaid shared savings methodologies through providers such as AMCs and CHCs.28 The CHC would contribute robust PCMHs that reduce preventable hospitalizations and readmissions for medically and socially complex Medicaid beneficiaries, which is essential to achieving cost savings in ACOs.29 The AMC would contribute greater financial resources and more experience managing the downside risk required in many shared savings models. By combining the best elements of AMCs and CHCs, Medicaid CHAMP ACOs could deliver high-quality, cost-effective care while training the next generation of health care professionals to effectively function in a group practice and a global payment environment. Controlling Medicaid costs will almost certainly be part of any long-term

Academic Medicine, Vol. 88, No. 12 / December 2013

federal budget agreement. The Center for Medicare and Medicaid Innovation (CMMI), funded at $1 billion per year over 10 years by the ACA, could expand its existing Medicare ACO models after determining whether CHAMP Medicare ACOs achieve cost savings and quality improvements for Medicaid beneficiaries. CMMI has prioritized new models of workforce development and deployment in previous initiatives such as the Health Care Innovation Awards. An initiative supporting CHAMP ACOs would pilot an innovative approach to preparing primary care physicians to practice in new care delivery models. The CHAMP Consortium Agreement

The CHAMP consortium agreement would create a formal, contractual partnership entity under joint governance of the CHC and the AMC; this entity would oversee the CHAMP THC “track” of each affiliated residency program (Figure 1). Such an entity, clearly recognized by the ACGME (for purpose of accreditation) and the Centers for Medicare and Medicaid Services (for the purpose of Medicare GME reimbursement), would constitute an academic consortium model.30,31 The CHAMP consortium would be designed to function under the existing ACGME accreditation that accredits both the sponsoring institution and individual programs within the AMC teaching hospital, overcoming one of the major administrative and economic challenges encountered in establishing current THCGME programs. The relationship between the CHAMP consortium and the AMC teaching hospital would be governed by the “Participating Institution” requirements of the ACGME, thereby precluding the need for independent institutional or program accreditation of the THC. The details of this arrangement would have to satisfy CHC federal funding mandates and would differ from the current THCGME program requirements. Program-level GME funding—both direct and indirect for inpatient training for the CHAMP track residents—would flow via the CHAMP consortium to the AMC where the THC residents receive their hospitalbased training (Figure 2). Similarly, the CHC will receive funds through the CHAMP consortium for outpatient training accomplished in the THC.

1839

Perspective

Figure 1  The community health center (CHC) and academic medicine partnership (CHAMP) consortium model’s accreditation process. AMC indicates academic medical center; GMEC, graduate medical education committee; ACGME, Accreditation Council for Graduate Medical Education. This figure depicts the maintenance of ACGME accreditation for the institution and for individual specialty programs by the AMC. The three shaded links to the THC represent the CHAMP program tracks, each of which supports four residents per year. The THC assumes administrative authority and responsibility for the THC portion of the educational effort. The CHAMP (CHC-AMC) Consortium, described in the text, is a joint entity involving the THC leadership and the AMC (teaching hospital), with links at both the program and institutional levels. THC leaders are members of the AMC GMEC. The linkage at both the program level and the GME-THC level assures that input, oversight, and governance responsibilities are shared for these tracks by both the AMC and the THC.

required residency programs would be accomplished by submitting requests to each of the respective ACGME Residency Review Committees to expand the AMC’s existing primary care residency programs and to approve the THC as a participating institution in the program. Second, the THC will not be required to create a duplicative administrative infrastructure for each accredited track, thereby minimizing the administrative and fiscal burden of the CHC and enhancing the feasibility of participation. Third, the time to approval by ACGME would be accomplished in a few months if not a few weeks, rather than in the 12 to 18 months sometimes required to approve a new THCGME program. Fourth, the flow of funds that we have described (Figure 2) would provide the CHC with sufficient leverage in its relationship with the AMC and its residency programs to preclude unilateral action by the AMC. This joint funding would foster truly collaborative and equal relationships. Finally, the model permits engagement of more than one AMC in a CHAMP consortium, which enables the involvement of an independent children’s hospital. Funding of CHAMP THCs

Governance of the CHAMP THC programs would reside within the consortium, a joint venture of the CHC and AMC, in which the participating CHCs would have an equal role as consortium partners. The consortium would have the obligation to distribute GME funds to the participants (the AMC and the CHC). Each of the three specialty residency programs would create their own CHAMP THC track, funded through the CHAMP consortium. Administrative and accreditation procedures would remain the responsibility of each of the three ACGME-accredited AMC residency programs (Figure 1), with required input from the consortium for only the CHAMP THC track. Each program would designate a CHAMP THC faculty member from the discipline to serve as both the track director and the THC site director (for oversight of resident activities at the CHC). The consortium agreement would also require the THC’s leaders to serve on the AMC’s GME committee. As noted, each primary care residency program would expand its resident complement by four funded positions per year, creating significant incentive for AMCs to participate and create a

1840

CHAMP THC track within each primary care program. The requirement that all three primary care residencies participate would need to be flexible because some CHCs may not have an adequate number of pediatric patients to support a pediatric residency program. Funding of these incremental positions would flow through the CHAMP consortium to both the AMC and CHC, as described, providing both incentive for the AMC to participate and leverage for the CHC to engage as an equal partner. To facilitate the formation of consortia, the ACGME could provide a template for a consortium agreement when an application for a CHAMP THC track was initiated. This model has clear benefits. First, the CHAMP consortium does not require that the THC independently achieve ACGME institutional accreditation, nor does it require creation of a newly accredited residency program in each specialty. Rather, institutional accreditation remains with the already-accredited AMC, and the CHAMP THC track would need to conform to the CHAMP consortium agreement. The establishment of CHAMP tracks in each of the three

The funding of GME costs for CHAMP THCs could be accomplished by a modification of the Centers for Medicare and Medicaid Services mechanism supporting the direct and indirect costs of GME. This modified funding mechanism would require an effective CHAMP THC consortium agreement, mandating the guidelines for governance and flow of funds necessary to ensure continued high-quality GME in both the CHC and AMC, which, in turn, would help ensure ACGME accreditation of the CHAMP THC tracks (Figure 1). For support of each participating AMC, the Centers for Medicare and Medicaid Services would direct to the consortium the appropriate annual per-resident amount of direct medical education funding, as well as the increment in indirect medical education payment related to the increased number of AMC THC track positions (Figure 2). The consortium would receive a perresident amount of $150,000 per year to support CHCs for the third year of ambulatory training, an amount similar to that established for the current THCGME program.10 The consortium agreement would require payment of all CHC GME costs, including those incurred in

Academic Medicine, Vol. 88, No. 12 / December 2013

Perspective

would support CHAMP training tracks with stable and sustainable long-term funding, thereby reassuring prospective residents and faculty members of long-term stability. The proposed flow of funds (Figure 2) would guarantee that both CHCs and AMCs would receive adequate support. Participating hospitals would receive their full direct and indirect medical education support per CHAMP THC resident from the consortium during inpatient training. Thus, previously capped hospitals would gain support for an increased number of residents and would require fewer hospitalists to staff services not covered by residents. Figure 2  The community health center (CHC) and academic medicine partnership (CHAMP) consortium model’s funding mechanism, with the internal medicine CHAMP teaching health center (THC) track shown as an example. Here the AMC (academic medical center or teaching hospital) would be accredited by the Accreditation Council for Graduate Medical Education (ACGME); the internal medicine program would also be accredited by the ACGME; and the internal medicine CHAMP THC track (and site) would be approved by the residency review committee. In a traditional funding mechanism, the Centers for Medicaid and Medicare (CMS) would send direct and indirect graduate medical education (GME) funds to the AMC for existing internal medicine residents. In the CHAMP THC funding mechanism, GME perresident funds from CMS flow to the CHAMP THC Consortium for internal medicine CHAMP THC residents. From the consortium, the funds (A) flow to the AMC for the direct and indirect GME expenses of hospital residency training. From the consortium, CHAMP THC (B) funding flows to the THC for the expenses of resident training for time spent in THC (see text for full details).

providing the residents’ THC continuity clinics2 during the first two years of their training; these costs would be covered by the AMC consortium support. Federal funding would depend on the enactment of currently proposed legislation. Two independent but similar House and Senate initiatives have a compelling rationale and bipartisan support. Both the “Training Tomorrow’s Doctors Today Act” (H.R. 1201) and the “Resident Physician Shortage Reduction Act of 2013” (S. 577) would phase in an additional 15,000 Medicare-supported residency positions over five years. Emphasis on training in community or outpatient settings is a priority in both bills. Support for 1,000 first-year CHAMP THC positions (of 3,000 to be established in 2015) would be consistent with the goals of this legislation. Enactment of this legislation would initiate funding for approximately 83 CHAMP THC programs, each composed of 12 residents (1,000 positions devoted to 83 programs of 12 residents).

of $30,000 for direct medical education funding, and additional indirect medical education support of $60,000 per CHAMP THC track resident, these 1,000 first-year positions located in 83 programs would require $90 million. Thus, a total AMC support of $180 million per year would be required for the first two years of inpatient training. If CHAMP THCs receive the same $150,000-per-resident amount for year three ambulatory training as the THCGME program,10 the total yearly cost of the 3,000 CHAMP THC positions when 83 three-year programs have been fully implemented would be approximately $330 million per year. If additional Medicare GME funding was unavailable, reallocating approximately 5% of the current $6.5 billion per year that funds Centers for Medicaid and Medicare Services indirect medical education would achieve budget neutrality. If Health Resources and Services Administration (HRSA) funding were maintained for pediatric GME, the Medicare costs could be reduced if the one-third of the residents who are pediatric trainees were supported by HRSA.

Assuming an average 30% Medicare inpatient load, thus a per-resident amount

Using the current Medicare GME funding mechanism in this manner

Academic Medicine, Vol. 88, No. 12 / December 2013

Conclusion

Linking ambulatory primary care GME with care of the underserved was first proposed more than 25 years ago.32 Family medicine pioneered the development of THCs, which have produced excellent outcomes in trainee satisfaction and in graduates’ pursuit of primary care careers, often focused on the underserved.9 The ACA’s THCGME program builds on this experience. It funds predominantly family medicine programs with varying levels of academic medicine partnerships and will undergo rigorous evaluation.10 We believe that the rapidly growing demand for health care requires rapid expansion of THCs to meet those health care needs and to benefit U.S. medicine. The various CHAMP THC features we have described support this expansion and can serve to complement the current THCGME program (Chart 1), which serves as an essential training venue for primary care in such areas as rural health, dentistry, psychiatry, and geriatrics. We have proposed a sustainable, longterm federal mechanism to fund a substantial number of CHAMP THCs. The stability of this mechanism,23 as well as numerous advantages for participating AMCs and CHCs, should provide incentive for developing CHAMP THCs. Given compelling evidence of THC effectiveness,2,9 and the urgency of expanding the supply of primary care physicians, we believe that the initiative to create CHAMP THCs should begin without delay. The program that we propose could quickly produce 1,000

1841

Perspective

Chart 1 Innovative Community Health Center and Academic Medicine Partnership (CHAMP) Teaching Health Center (THC) Features Compared With the Affordable Care Act–Funded Teaching Health Center Graduate Medical Education (THCGME) Program CHAMP THC

THCGME Program10

Provides concurrent residency training for three primary care specialties (family medicine, pediatrics, internal medicine) Develops a specific clinical role for each primary care discipline while integrating all three

Is predominantly a family medicine residency training model, but supports dental, geriatric, mental health, and obstetrics–gynecology training Offers an extensive clinical role for residents in predominantly one primary care discipline

Includes in the residency curriculum interdisciplinary and multidisciplinary education with emphasis on patient-centered medical home practices and use of electronic medical records

Includes in the residency curriculum delivery models that support the provision of contemporary primary care services, including electronic medical records, interdisciplinary care teams, and management/leadership training

Develops virtual multispecialty group practice, facilitating subspecialty consultations, as well as accountable care organization and other potential cost-saving innovations

Has limited potential for cost-saving innovations and facilitated subspecialty consultation

Is based on a well-conceived and executed CHAMP consortium agreement that provides an effective governance structure, maintains community health center (CHC) culture, ensures generous funding, and facilitates residency accreditation

Works under variable affiliation agreements with academic medical centers, features variable THC administrative structures, and contends with some accreditation challenges due to the CHC “operating the residency program”

Creates a Centers for Medicaid and Medicare mechanism for stable, long-term funding of primary care graduate medical education expansion as part of currently proposed legislation

The Affordable Care Act mandates $230 million funding for only five years via Health Resources Services Administration budget; requires reauthorization in 2015. Uncertain funding is current barrier to program expansion.

Has a projected output of 1,000 primary care physicians per year if 83 twelve-resident programs are established

Has a projected output of 40 primary care physicians per year from the 11 inaugural programs established in 2011

additional primary care physicians per year. More innovative CHAMP THC features, such as the potential for Medicaid ACOs, could be tested with the support of the CMMI. CHAMP THCs have the potential to reshape primary care GME, attracting more students into primary care residency programs and preparing an expanded cadre of residents for practice in the 21st century. Although medical school enrollment is on track to reach a 30% increase by 2016,33 this increase will not substantially impact the nation’s primary care workforce shortage without a concomitant increase in primary care residency positions. The number of federally supported residency training positions has remained essentially unchanged for 15 years since Congress enacted a freeze as part of the Balanced Budget Act of 1997.34 Thus, currently proposed GME expansion legislation is an important initiative. Its inclusion of 1,000 CHAMP THC positions per year would be an excellent mechanism to target a compelling societal need. Participation in a CHAMP provides AMCs with the opportunity to partner in an essential societal mission—rebuilding our nation’s primary care infrastructure. We describe how the CHAMP THC would enable rapid expansion of primary care GME in a manner designed to prepare physicians for practice in the 21st

1842

century. We also describe accreditation and funding mechanisms that would facilitate development and expansion of the current THCGME model. Properly implemented, CHAMP THCs can effectively reverse the growing shortfall in primary care providers, particularly for low-income and vulnerable populations. Acknowledgments: The authors wish to thank Drs. David Feldstein, Arthur Kellerman, Kenneth Loving, and Robert Phillips for their sucggestions and contributions, and Ms. Kathy Holland for her invaluable assistance in preparing this manuscript.

Other disclosures: None. Disclaimer: Dr. Rieselbach and Dr. Nasca are past presidents of the Association of Program Directors in Internal Medicine. This proposal does not necessarily reflect the position of the foregoing organization. Ethical approval: Not applicable. Dr. Rieselbach is professor emeritus of medicine, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin.

Dr. Neuhausen is director, Delivery System Reform, Office of Health Innovation, and clinical assistant professor, Department of Family Medicine and Population Health, Virginia Commonwealth University, Richmond, Virginia.

Dr. Frohna is professor of pediatrics, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin, and program director, American Family Children’s Hospital Pediatric Residency, Madison, Wisconsin.

References

Funding/Support: None.

Dr. Crouse is professor of family medicine, associate dean, Rural and Community Health, and director, Wisconsin Academy of Rural Medicine, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin.

Dr. Nasca is chief executive officer, Accreditation Council for Graduate Medical Education, Chicago, Illinois, and professor of medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania.

1 Starfield B. The future of primary care: Refocusing the system. N Engl J Med. 2008;359:2087, 2091. 2 Rieselbach RE, Crouse BJ, Frohna JG. Teaching primary care in community health centers: Addressing the workforce crisis for the underserved. Ann Intern Med. 2010;152:118–122. 3 Institute of Medicine. Committee to Study Strategies for Supporting Graduate Medical Education for Primary Care Physicians in Ambulatory Settings. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: National Academy Press; 1989. 4 Adashi EY, Geiger HJ, Fine MD. Health care reform and primary care—the growing importance of the community health center. N Engl J Med. 2010;362:2047–2050. 5 Rosenbaum S. Reinventing a classic: Community health centers and the newly insured. In: Hall MA, Rosenbaum S, eds. The Health Care Safety Net in a PostReform World. New Brunswick, NJ: Rutgers University Press; 2012:67–90. 6 Kaiser Family Foundation. Community Health Centers in an Era of Health Reform: An Overview and Key Challenges to Health Center Growth. March 2013. http://kff. org/health-reform/issue-brief/community-

Academic Medicine, Vol. 88, No. 12 / December 2013

Perspective

7

8

9

10 11

12

13

14

15

health-centers-in-an-era-of-health-reformoverview/. Accessed August 27, 2013. Neuhausen K, Grumbach K, Bazemore A, Phillips RL. Integrating community health centers into organized delivery systems can improve access to subspecialty care. Health Aff (Millwood). 2012;31:1708–1716. Rieselbach RE, Remington PL, Drezner MK, Golden RN. Expanded community health center–academic medical center partnerships. WMJ. 2011;110:168–169. Morris CG, Johnson B, Kim S, Chen F. Training family physicians in community health centers: A health workforce solution. Ann Fam Med. 2012;10:163–168. Chen C, Chen F, Mullan F. Teaching health centers: A new paradigm in graduate medical education. Acad Med. 2012;87:1752–1756. Oliver TR, Rieselbach RE, Crouse BJ, Frohna JG. Community health center and academic medical partnerships to expand and improve primary care. In: Hall MA, Rosenbaum S, eds. The Health Care Safety Net in a PostReform World. New Brunswick, NJ: Rutgers University Press; 2012:108–125. Mullan F, Chen C, Petterson S, Kolsky G, Spagnola M. The social mission of medical education: Ranking the schools. Ann Intern Med. 2010;152:804–811. Ku L, Jones E, Finnegan B, Sin P, Rosenbaum S. How Is the Primary Care Safety Net Faring in Massachusetts? Community Health Centers in the Midst of Health Reform. March 2009. http://kaiserfamilyfoundation. files.wordpress.com/2013/01/7878.pdf. Accessed August 27, 2013. Kellermann AL, Weinick RM. Emergency departments, Medicaid costs, and access to primary care—understanding the link. N Engl J Med. 2012;366:2141–2143. Macy Foundation. Ensuring an Effective Physician Workforce for the United States: Recommendations for Reforming Graduate Medical Education to Meet the Needs of the Public. May 2011. http://www.

16

17 18

19

20

21

22 23

24

Academic Medicine, Vol. 88, No. 12 / December 2013

macyfoundation.org/docs/macy_pubs/ Macy_GME_Report,_Aug_2011.pdf. Accessed August 27, 2013. Macy Foundation. Ensuring an Effective Physician Workforce for America: Recommendations for an Accountable Graduate Medical Education System. October 2010. http://www.macyfoundation. org/docs/macy_pubs/JMF_AAHC_ Summary_2_20111.pdf. Accessed August 27, 2013. Golden AG, Silverman MA, Mintzer MJ. Is geriatric medicine terminally ill? Ann Intern Med. 2012;156:654–656. Leipzig RM, Hall WJ, Fried LP. Treating our societal scotoma: The case for investing in geriatrics, our nation’s future, and our patients. Ann Intern Med. 2012;156:657–659. Relman AS. How doctors could rescue health care. N Y Rev Books. October 27, 2011. http://www.nybooks.com/articles/ archives/2011/oct/27/how-doctors-couldrescue-health-care/. Accessed July 27, 2013. Willard R, Bodenheimer T. The Building Blocks of High-Performing Primary Care: Lessons From the Field. April 2012. http:// www.chcf.org/publications/2012/04/ building-blocks-primary-care. Accessed August 27, 2013. Nasca TJ, Philibert I, Brigham T, Flynn TC. The next GME accreditation system— rationale and benefits. N Engl J Med. 2012;366:1051–1056. National Health Service Corps Web site. http://nhsc.hrsa.gov/. Accessed August 27, 2013. National Health Service Corps. Students to Service Program. http://nhsc.hrsa.gov/ loanrepayment/studentstoserviceprogram. Accessed August 27, 2013. National Health Service Corps. Loan Repayment Program FAQs. http:// nhsc.hrsa.gov/currentmembers/ loanrepaymentrecipients/faqs/. Accessed August 27, 2013.

25 Reynolds PP. Commentary: The teaching community health center: An idea whose time has come. Acad Med. 2012;87:1648–1650. 26 Rieselbach RE, Kellermann AL. A model health care delivery system for Medicaid. N Engl J Med. 2011;364:2476–2478. 27 Berwick DM. Making good on ACOs’ promise—the final rule for the Medicare shared savings program. N Engl J Med. 2011;365:1753–1756. 28 Centers for Medicare and Medicaid Services. Policy considerations for integrated care models. Letter to state Medicaid directors. July 10, 2012. http://www.medicaid.gov/ Federal-Policy-Guidance/downloads/SMD12-002.pdf. Accessed August 27, 2013. 29 Colla CH, Wennberg DE, Meara E, et al. Spending differences associated with the Medicare Physician Group Practice Demonstration. JAMA. 2012;308: 1015–1023. 30 Accreditation Council for Graduate Medical Education. Teaching Health Center Graduate Medical Education (THCGME) Initiative. August 2012. http://www.acgme.org/ acgmeweb/Portals/0/THCGME_FAQ.pdf. Accessed August 27, 2013. 31 Balanced Budget Act of 1997. Pub L No. 105– 33, 111 Stat 484. Section 4628: Demonstration project on use of consortia. 1997. 32 Rieselbach RE, Jackson TC. In support of a linkage between the funding of graduate medical education and care of the indigent. N Engl J Med. 1986;314:32–35. 33 Association of American Medical Colleges. The Impact of Health Care Reform on the Future Supply and Demand for Physicians Updated Projections Through 2025. June 2010. https://www.aamc. org/download/158076/data/updated_ projections_through_2025.pdf. Accessed August 27, 2013. 34 Iglehart JK. The residency mismatch. N Engl J Med. 2013;369:297–299.

1843