Amidst the debate about health care reform, there appears to be near unanimity around the fact that a reformed US health care system requires at its foundation a robust system of primary care. The Patient-Centered Medical Home (PCMH) has emerged as the leading strategy around which primary care will be redesigned.1 The core principles of the PCMH model build upon the core concepts of primary care as defined by Starfield2 and the Institute of Medicine3, and include a whole person orientation with care that is accessible, coordinated, comprehensive, and continuous over time.

There are a variety of challenges to implementing the PCMH model; the core principles serve as a general guide, but do not necessarily specify the required capabilities of PCMH practices, the optimal reimbursement strategy, or the ideal methods for facilitating the transformation of current practices to meet the ideals of the PCMH model of care. Thus, although the implementation of the PCMH should be grounded in an evidence base supported by scientific research, new research must accompany policy development so as to inform the optimal implementation of the PCMH and track the effects of the PCMH on care delivery.

Recently, the three major physician-oriented academic primary care societies [the Society of General Internal Medicine (SGIM), the Society of Teachers of Family Medicine (STFM), and the Academic Pediatric Association (APA)] brought together experts in primary care, health services and implementation research, health systems and insurance, and policy makers to develop a policy-relevant research agenda for the PCMH. The conference included a series of commissioned papers developing research questions and exploring implementation challenges to various aspects of the PCMH model whose themes were selected by an expert steering committee. In addition, the conference also convened a panel of expert policy makers to react to a prioritized list of research questions that was developed for the conference. The series of commissioned papers, reflecting the discussions that occurred at the invitational conference, are published together as a special symposium in this issue of the Journal of General Internal Medicine.

The primary purpose of the conference was to develop a policy-relevant research agenda for the PCMH. The conference and papers highlight the emerging popularity of this model of practice. As shown in the article by Bitton et al.,4 26 demonstrations were active by the end of 2009, and a further 65 or more demonstrations are under development. Countless other practices or systems likely are experimenting with this model, even in the absence of external payment reform. Although previous research shows that areas of the country and systems of care built upon a model of enhanced primary care have better performance,5 little research has demonstrated the feasibility or effectiveness of implementing the PCMH within the context of the current US health care system. Thus, a carefully thought out agenda of research needs must accompany the implementation of this model in the US if we are maximize the effectiveness and assure the continuation of the primary care function in the future.

The conference and resulting papers are a significant step in framing that research agenda, and are notable for a number of reasons. First, this effort represents what we hope will become an ongoing collaboration between SGIM, STFM, and APA, the principal academic primary care societies. The collaboration is built upon the premise that the many commonalities between the primary care disciplines far outweigh any differences or perceived competition for scarce resources. Thus, the shared values and goals across the specialties serve as a foundation from which to advocate for and advance the science and practice of primary care and generalism. Such collaboration should not only enhance the voice of primary care scholars in policy circles, but also facilitate learning across the specialties, and foster innovative partnerships and research across the disciplines of primary care that will inform future practice and the implementation of the PCMH.

Second, although there is substantial enthusiasm for reforming primary care in general, and the PCMH model of care in particular, there are many unanswered questions that need to be addressed. A tension exists between researchers (who seek well-designed experiments to maximize learning and define optimally the components and paths to implementation of the PCMH), primary care advocates (who are eager to secure enhanced support for a model of care they judge of proven value), and pragmatic implementers (who seek to accelerate the broad implementation of this model, but who also understand that learning and adjustment will be necessary over time). Moreover, this debate is heavily influenced by policymaker concerns about unsustainable health care costs and the severe limitations of the US system of primary care. Current projections foresee a conflict between the growing need for adult primary care services and a shrinking number of graduating medical students and residents entering primary care to replenish the aging population of current primary care practitioners. In addition, as private and government purchasers of care face ever increasing health insurance costs, they look towards the PCMH as a method for controlling rising expenditures. They need quick solutions, not multi-year experiments, however, and will have difficulty supporting a model without some expectation of potential savings.

Third, the set of papers that resulted from the conference further explores some of the challenges associated with implementing the PCMH model and the important research questions that should be answered to inform policy makers eager to move ahead with successful reform. Together, the articles highlight the many challenges that policy makers and implementers will need to overcome as we move to implement this model of care. Bitton et al. provide data from a national survey of all currently active PCMH demonstration programs that include external payment reform.4 The paper highlights the diversity of approaches being taken within the current demonstrations, as well as areas of commonality that will facilitate comparisons. Notably, current demonstrations include over 14,000 physicians caring for nearly 5 million patients. The paper by Stange et al.6 defines the PCMH as a merger of the fundamental components of primary care with new ways of organizing and enhancing practice, and changes in reimbursement to support this model. The authors argue that measurement of the PCMH components is critical to help transform practices as well as guide appropriate reimbursement. They suggest that current measurement tools are limited by their overemphasis of the technical PCMH components and underemphasis of core primary care components, but that merging parts of different tools may help to address those limitations.

Homer and Baron address the issue of practice transformation.7 They note that such transformation goes beyond small-scale changes within a practice and will require a profound realignment of the roles and responsibilities within primary care teams that focuses on value creation for patients and the delivery system overall. Simply changing payment and establishing learning collaboratives may not be sufficient to drive behavior change over the short time horizons of most demonstrations so methods will need to be refined to facilitate practice transformation. In their first paper, Berenson and Rich explore in depth the discouraging history of inadequate payment for primary care in the US, with their second paper offering a detailed analysis of payment reform options to reimburse medical home services.8,9 These two papers highlight the policy challenges of designing optimal payment incentives for the PCMH and suggest there may be no single perfect system. Instead, PCMH payment reform will need to accommodate real-world feasibility, with an eye to the different needs of patients and communities, as well as the varying configuration and resources of clinical practices. Rittenhouse and colleagues10 explore the outcomes that are likely to result from implementing the PCMH model at the levels of individual patients, practices, and communities. Finally, Pham11 explores how patients and clinicians might interact in a medical neighborhood and potential policies for promoting such integration and coordination across the continuum of care.

The discussions at the conference also highlighted additional areas of importance. Although the focus of the conference was the policy research agenda relevant to making the medical home a reality in typical community practices, medical education is a key mission of all three of our academic primary care associations. The PCMH is likely to build upon team-based models that utilize a variety of additional personnel such as physician assistants and advanced practice nurses, other nursing professionals, social workers, and health educators to meet diverse patient needs for accessible and comprehensive patient-centered care. Within these new teams the critical over-arching function of care coordination must be fulfilled as well. Thus, we have much to learn about who should constitute the members of these teams, the roles of various team members, and how best to prepare future primary care physicians to practice within such teams. Over the past year, primary care policy advocates from our societies have worked with Congressional staff and others to help craft relevant federal training initiatives in the context of health care reform. Once the future policy context of the medical home has been established, there is much more work to be done on how best to train the next generation of primary care physicians.

Thus, although the initial implementation of the PCMH is grounded in a substantive evidence base, the papers in this JGIM symposium suggest that new research must accompany policy development so as to refine the implementation of the PCMH and optimize the benefits of the PCMH on care delivery. The questions and research agenda raised in the accompanying papers go a long way towards defining what needs to be answered in both the near and medium term to reform US primary care. However, optimal implementation of the PCMH will require continuous learning from thoughtful policy-relevant research, not just of ongoing regional demonstrations, but ultimately from nationwide payment reform programs (like the “pilots” in health care reform) with the scale and scope to guide broad-based transformation of primary care practice.