Context for Understanding the National Demonstration Project and the Patient-Centered Medical Home

I strongly agree with Dr. Iliff's last post, and feel it is deserving of a response.

The absence of financial data in the NDP evaluation puts the entire project into the "interesting, but irrelevant" category for practicing physicians. If the participating practices were initially unable to generate simple financial statements, something was wrong with the selection process. If, after two years, the practices and facilitators were still unable to come up with the needed data, there was something wrong with the way the project was conducted (and I certainly wouldn't look to these practices and facilitators for advice on practice management!). And, if the NDP still refused to present the requested information to the AAFP at the conclusion of the project, the AAFP should have rejected the findings outright. Instead, the AAFP stated (somewhat tautologically) that "the TransforMED NDP achieved what it set out to do," and proceeded to ignore the issues raised in the final reports.

I think the lack of postings here is indicative of the general disappointment family physicians feel about the manner in which the NDP results have been handled.

Competing interests:   None declared

We write to provide a North Carolina perspective on the methodology of the National Demonstration Project.

We are active participants in a series of large scale interventions to develop new models of practice and spread them throughout a whole state. The foundation is Community Care of North Carolina (CCNC)[1-3], a statewide Medicaid network which includes over 1200 practices organized in 14 regional networks. CCNC has demonstrated significant improvements in quality and cost effectiveness[3] and is one of the major national models for health care reform. In addition, building on CCNC and the infrastructure of the North Carolina AHEC program, we have developed the North Carolina Improving Performance in Practice project (NC IPIP) which started as a national initiative through the American Board of Medical Specialties funded by Robert Wood Johnson Foundation. NC IPIP is now a part of our newly developed North Carolina Health Quality Alliance (NCHQA)[4], which includes specialty societies, all major North Carolina payers, state government, employers and other health care organizations. The goal of NC IPIP/NCHQA is to improve dramatically the quality of care in all 2000 primary care practices in North Carolina. The results of our first wave of 18 practices have just been published[5], and we are now engaged with over 170 practices. Led by NC AHEC, the work of NC IPIP is now expanding to over 750 practices that receive services via the ONC funded HIT Regional Extension Center program. Finally, along with colleagues in South Carolina and Virginia, we have led two large scale collaboratives to improve quality of care in residency practices: the I3 collaborative in 10 Family Medicine residencies, and now, the I3 PCMH Collaborative, involving 25 residencies (including Family Medicine, Internal Medicine and Pediatrics) in NC, SC and Virginia who are working to obtain PCMH recognition. We have demonstrated significant improvement in measured quality of care in Diabetes and CHF, including a 25% reduction in hospitalizations for CHF, and are making excellent progress in helping residencies across the primary care specialties gain PCMH recognition.

The publication of the National Demonstration Project is a milestone in the life course of Family Medicine. The many strengths of the National Demonstration Project are manifest. The patient centered medical home (PCMH) model tested comes directly out of the construct developed in the Future of Family Medicine project and the joint statement. The evaluation was prospective, with concurrent data collection, and there was a substantial effort to separate evaluation from the project itself. Grounded in an understanding of practices as complex adaptive systems, the project explicitly acknowledged that the intervention was evolving; appropriately, and courageously, the evaluators asked whether practice facilitation helped with practice transformation. Finally, and of particular merit, the evaluation used a multimodal approach pioneered over the last 15 years by members of the evaluation team. A combined quantitative and qualitative evaluation is appropriate and recognizes the need to capture outcomes along many different dimensions.

Within this context, we would like to raise several issues for dialogue. The first is generalizability. The NDP practices are a highly selected, highly motivated set of practices. The high penetration of EHRs in both the facilitated and the non-facilitated groups, as well as the high degree of PCMH features at baseline confirm that these practices were quite different from the usual primary care practice. We would like more information about the initial selection from 337 practices to the final 36, as well as more information about the functional aspects of the practices at baseline—financial footing as well as standing orders, registries, and other aspects of organizational function such as practice meetings. The lack of general internal medicine and pediatric practices may also limit the generalizability of the findings to all primary care.

The NDP practices are clearly "early adopters."[6] From the perspective of our large scale interventions, we must be concerned with all practices, not just the early adopters, and lessons learned with early adopters may not hold true for the "early majority" or even later adopters. For example, in NC IPIP/NCHQA, we have learned that early adopters crave new ideas and thrive on change—money and other material benefits seem less important. We believe later adopters will require more incentives to transform which may include payment, CME and Maintenance of Certification credit, for example. This early adopter bias is often unrecognized and common to most early trials of all clinical innovations.

A second major issue is the robustness of the intervention. We often describe a spectrum of intensity of interventions with practices in our various settings as ranging from Yaris to Corolla to Camry to Lexus. Judged against the yardstick of our experience, the NDP was a Camry. The report describes facilitators with experience, who spent substantial time with the practices initially and through 4-6 site visits, and had substantial phone, email and other contact with the practices; in addition, four face to face learning sessions facilitated learning across practices.

We would like more information about the selection, training and quality improvement process for the facilitators. In NC IPIP/NCHQA, we currently have 12 Consultants (QICs), and our major criterion for selecting QICs has become prior experience in quality improvement projects – and work in hospital QI is sufficient. Beyond experience with QI, we have found that the key trait of QICs is the ability to develop and maintain a constructive relationship with individual practices over time and to adapt to the style and needs of the practice. Over four years, we have worked hard to develop explicit processes for not only sharing learnings across practices, but also standards for training and evaluating the performance of QICs.

The intensity and quality of NDP practice facilitation is unclear. In terms of the ratio of facilitators to practices, the NDP used a 1:6 ratio. In practice redesign projects in North Carolina before NC IPIP started, the range of ratios was between 1:3 and 1:100; after four years of experience with NC IPIP, and with increasing experience of QICs in the field, we currently think the optimal ratio is 1:20-25, though limitations of numbers of QICs and increased demand may force our actual ratio higher. Another key element of practice facilitation is the number and kind of contacts with the practice. In the NDP, the facilitators spent a substantial amount of time on site at the beginning and in site visits every 4-6 months, along with emails and other contact and learning sessions twice a year. In contrast, taking advantage of their regional placement, NC QICs visited practices in person twice a month or more often initially in addition to email and other contacts and organized regional learning sessions quarterly. Finally, perhaps the most important potential weakness of the NDP is the relative rarity of organized data feedback to the practices and relatively modest support for the learning network among practices. In both NC IPIP, as well as the I3 and I3 PCMH Collaboratives, we have fed back performance data to the practice on a monthly basis, and this has proved to be a powerful enabler of change. In addition, from day one, we have emphasized learning lessons from other practices; quarterly regional collaborative meetings have played a key role in developing local learning networks. We also invite other clinicians, nurses and all other staff to the regional collaborative meetings—not just the clinical champion and the practice manager.

How much of practice redesign requires local community based support and how much can be done at a distance? Our reading of the recent Medicare demonstration projects on care management suggests personal relationships are important—that regular, direct person-to-person contact between either doctor and patient, or case manager and patient is critical for improving outcomes—is this also true for facilitation of practice redesign?

A final issue is measurement. The NDP importantly attempted to measure the experience of the patient in the core functions of primary care. Although response rates were disappointing, the attempt to ascertain patients' experience is a strength of the study. The attempt to measure quality was also appropriate, as was the choice of a set derived from a national standard, the AQA starter set. The NDP's use of a convenience sample of patients is likely to be more sensitive to change but less valid than the population measures we have used in NC IPIP. While convenience sampling is often more sensitive to change, we believe that estimates of the whole population either through random samples or an EHR total pull, should be preferred. In any case, the choice of sampling strategy needs to be explicit.

Which diseases to track in terms of quality remains an important question. The premise of the AQA starter set is that, ultimately, primary care practices will need to improve care for many different diseases. By contrast, given our focus on practice redesign for quality improvement with chronic disease, the I3 Collaborative as well as NC IPIP have focused on specific diseases (Diabetes/CHF and Diabetes/Asthma, respectively), at least initially. We currently believe that practices need to learn how to transform care for one disease initially and then spread to other diseases. Moreover, as we address cost control across the whole state North Carolina Medicaid population in CCNC, we believe increasingly that measurement of single diseases alone is inadequate to capture the highest risk/highest cost patients, and we are beginning to develop measures of care that integrate across diseases/comorbidities. We believe that once we get better measures—and EHRs that can deal with them—we will have a more realistic view of primary care’s holistic integrative function.

We would also like more information from the NDP about other major influences on practice transformation—baseline financial status of the practice, EHR implementation status, practice organization and related community initiatives. Many primary care practices in our region are on the brink of bankruptcy—and some aspects of practice redesign, including EHRs and chronic disease care can add substantially to the financial burden. We need to know where the NDP practices started from and where they ended. Does the medicine kill the patient? Similarly, EHR implementation impacts every part of practice operations, and it is important to know what exactly has been implemented, how it is being used, and what the capacity is of the software selected. Finally, the lack of consideration of a community component—hospitals/integrated health systems, public health organizations, community case managers—in the design and the discussion of the NDP is striking. Community remains an acknowledged, but often unstudied, component of the Chronic Care model[7]. One of the major lessons of CCNC, NC AHEC, NC-IPIP and NCHQA is the enormous impact that community based initiatives can have on practice redesign efforts—either positive, as in providing support for patient self-management, or negative as in distraction with different projects and different measures. Alignment is critical, never-ending—and very difficult.

We are immensely grateful to the practices and evaluators of the National Demonstration Project. We have learned a great deal from your vision, courage, and incredibly hard work!

Warren Newton, MD MPH, Dept of Family Medicine, University of North Carolina at Chapel Hill. Ann Lefebvre, MSW, Dept of Family Medicine, North Carolina AHEC, UNC Chapel Hill. Elizabeth Baxley, MD, Dept of Family and Preventive Medicine. University of South Carolina School of Medicine. Beat Steiner, MD MPH, Dept of Family Medicine, UNC Chapel Hill. Tom Bacon, Dr PH, NC AHEC, UNC Chapel Hill. Katrina Donahue, MD MPH, Dept of Family Medicine, UNC Chapel Hill. Samuel Weir, MD, Dept of Family Medicine, UNC Chapel Hill.

References

1. Community Care of North Carolina. [cited 2010 June 25]; Available from: http://www.communitycarenc.com/.
2. Willson, C.F., Building primary care medical homes within the community care of North Carolina program. N C Med J, 2009. 70(3): p. 231-6.
3. Steiner, B.D., et al., Community care of North Carolina: improving care through community health networks. Ann Fam Med, 2008. 6(4): p. 361-7.
4. NCHQA. [cited 2010 9/20/10]; Available from: http://www.ncquality.org/.
5. Newton, W.P., et al., Infrastructure for large-scale quality-improvement projects: early lessons from North Carolina Improving Performance in Practice. J Contin Educ Health Prof, 2010. 30(2): p. 106-13.
6. Rogers, E.M., Diffusion of innovations. 1962, New York, NY: Free Press.
7. Wagner, E.H., B.T. Austin, and M. Von Korff, Organizing care for patients with chronic illness. Milbank Q, 1996. 74(4): p. 511-44.

Competing interests:   None declared

As a fellow family physician sharing the same goals and rowing the same boat, I want to take the NDP team seriously, and I want to give you the benefit of the doubt.

Here's my best effort. I'm going to confine my comments to the section titled "Where is the Financial Data", which is not only the most important subject of the NDP, but also the area where we all seem to agree.

In my letter to Family Practice Management in February of 2008 (http://www.aafp.org/fpm/2008/0200/p8.html), and in my FPM blog post of November 14, 2008 (http://blogs.aafp.org/fpm/makingit/entry/productivity), I addressed this issue in detail. I recognize the extreme difficulty of aggregating financial data from different practices; it is an accounting nightmare.

I produced a simple worksheet which required nothing more than a few commonsense adjustments to the 1040 which every working American has to produce for the IRS every year (http://www.aafp.org/fpm/2008/0200/fpm20080200p8-rt1.doc). The calculation gives an "hourly income" figure, which is to my mind the only way to make an apples-to-apples comparison among physicians and practices over time. The worksheet is copyrighted by the AAFP and still available on the Academy website.

Therefore I'm not buying the authors' hand-wringing, at least until they explain why family physicians who voluntarily sign up for a multimillion dollar Academy experiment can't be asked to provide the tax return data which they have to produce every time they ask a bank for a loan. The 1040 data is transparent and, on pain of penalty, presumably honest. It reflects the only true "bottom line" that prospective family physicians really care about: how much money did you take home to the spouse and kids?

Furthermore, since the NDP physicians have to maintain these records for five years, they are still available for analysis.

Maybe I've got a screw loose somewhere, but until you explain: What's holding you up?

Competing interests:   None declared

As members of the evaluation team of the NDP, we write to offer our response and reaction to comments posted on the entire supplement attached to this link. There are multiple areas dealt with in the comments that we will try to address as headlines below.

THE CURRENT FEE-FOR-SERVICE, VOLUME-DRIVEN PAYMENT SYSTEM IMPEDES THE DEVELOPMENT OF THE PCMH

We agree with Drs. Iliff (6/11/10), Watkins (6/21/10), Bauer (6/28/10), and Green (7/2/10) that building a PCMH in the current environment, with its misaligned incentives and undercapitalization of primary care, undermine the development of the PCMH. As Dr. Iliff (6/27/10) points out "earning a better income is necessary but not sufficient for effective transformation." The current perverse incentives for volume undermine both the principles and the goals of the PCMH. However, the fact that we are in transition towards a better-aligned system should not lead to paralysis and acceptance of the status quo. Practices can work to increase their adaptive reserve and to better integrate with their local environment. (1,2) As Dr. Green (7/2/10) reminds us "fixing primary care with the ongoing evolution of the PCMH will not be possible…until primary care is capitalized."

The NDP attempted a financial analysis and found that most independent practices did not have sufficiently accessible financial data to accomplish this. This in itself is an important finding (3, pS13) and points to the need to develop the internal capabilities of primary care practices (4).

WHERE IS THE FINANCIAL DATA?

Dr. Iliff (6/8/10, 6/11/10, 6/27/10) and Dr. Watkins (6/21/10) emphatically express distress at the failure to adequately assess the financial impact of the NDP efforts. We couldn't agree more! The AAFP specifically asked for this financial data and both Transformed and the NDP evaluation team designed the project to obtain such data. The problems and ultimate failure to get meaningful before and after financial data reflect much of what is broken in our fragmented non-system of health care. Each of the practices had a different mix of payers making it impossible to negotiate their sharing of data within the timeframe of the study and within the constraints of institutional review boards. Each practice had a different organizational structure with unique, often complex, financial governance and, frequently, very different ways of collecting and organizing their financial information. To accommodate all of this variation, we developed a simple monthly accounting sheet for each practice to complete. Practices were unable to do that so we went to quarterly and offered assistance in doing the form. It was still not possible. Thus, the only meaningful comparative we have is the completed, simplistic financial overview at the end of the study which suggests that a few practices may have improved their income flow but many struggled. Several members of the NDP evaluation team are more than academics and also serve as leaders in managing practice networks and painfully aware of the daily struggles to maintain incomes, cash flow, and the ability to meet payroll. The absence of adequate financial data in the NDP IS DATA! There is a critical need to build practice core and financial management, and there is a greater need for sharing of financial data across our health system. Hopefully, the development of accountable care organizations will facilitate this.

A PCMH IS MORE THAN THE SUM OF ITS PARTS

Drs. Bauer (6/28/10) and Green (7/2/10) present us with important aspects of a definition of a true PCMH by emphasizing the whole practice transformation, i.e., involving everyone in the practice, and the fact that the PCMH is not an object but a journey towards a high functioning primary care practice with the patient at the center of changes. We agree with this assertion as we recently published our definition of the PCMH. (5)

Both Drs. Kuzel (6/8/10) and Green (7/2/10) emphasize the need for a new "mental model", a different approach, a personal transformation in the physicians before continuing in the journey. Drs. Lesko (6/17/10) and Green (7/2/10) remind us that a "personal physician" is still needed for highly efficient and effective primary care. We strongly agree with the value and need for a personal physician (6). While maintaining the need for this personal physician, Dr. Green (7/2/10) also calls us to:" [m]ove now from front and back offices with an adjudicating overlord called a physician to integrated systems of care with stunning, reliable teamwork, inclusive of mental health and linkage to public health."

Nevertheless, we take issue with the assertion from the 4 primary care organizations that a PCMH always needs to be "physician led" (2, pS86; 7). While the NDP provides abundant evidence of the importance of appropriate physician leadership, physicians are not the only ones who can or should provide leadership for renewed primary care. Abundant resources can and are being released by engaging diverse leadership in the PCMH (8)

TRANSFORMATIONAL CHANGE NOT INCREMENTAL CHANGE IS NEEDED FOR THE FUTURE

Medical student Margolius (6/19/10) reminds us of the perilous condition of primary care and the need for a major quantum leap. We agree with his assessment and plan. We also agree with Drs. Borkan (7/1/10) and Moore (6/11/10) that many of the current demonstration projects going on across the country are too narrow in their focus and have misaligned incentives. For the PCMH to be implemented effectively there is a need for "whole system" transformation, not only involving all in the practice but also incorporating the practice into the community systems (whether other health care systems or other community resources that can benefit patients). Only with wholesale transformation can the improvement in population outcomes and costs be seen as pointed out by Dr. Moore (6/11/10).

THE NDP, BOTH A DEMONSTRATION PROJECT AND A LEARNING LABORATORY

Dr. Watkins (6/21/10) questions the assertion that the NDP was never a demonstration project. We believe that the NDP was BOTH a demonstration project AND a learning laboratory. Our findings have already changed the dialogue about the road to improve primary care in our country (9). As mentioned above, we now know that we need to go beyond more payment to a change in the reimbursement models; we now know that we need to emphasize the 4 pillars of primary care (first contact access, comprehensive care, coordinated care, and personal relationship over time); we now know that the patient experience is critical for planning and patients must be included in the process of practice transformation and that practices need to be integrated in their healthcare neighborhood. Finally we now also know that change to a PCMH is hard and requires significant influx of resources (facilitation, dollars, change in mental models, and leadership development). As summarized by Dr. Sinsky (6/17/10) "The rich qualitative and quantitative results outlined in these eight articles mark a significant movement forward in our understanding of the work that will be necessary if we are to revitalize primary care in the U.S."

SEVERAL METHODOLOGICAL LIMITATIONS AND FUTURE CHALLENGES EMERGE FROM THE NDP EVALUATION

We discussed the study methodological limitations in each of the papers that were presented in the supplement. We share Dr. Sinsky's (6/17/10) important challenge to build more "robust tools" for measuring the four pillars of primary care and are actively involved in promoting changes in the NCQA process to incorporate measures of these pillars in their recognition process. We also acknowledge that for patient surveys to be more representative a higher response rate is necessary. With Dr. Antonucci (6/13/10) we recognize that our research measurement tools may be too cumbersome to be used routinely in evaluations of primary care practices. It is important to remember that patient surveys only measured patient experience up to 26 months from the beginning of a process that in some cases may take 3-6 years to complete. With Dr. Pratt (8/30/10), we recognize that a study with a passive control would have added information, but both cost and feasibility made a third group not doable. The practices that applied to be in the NDP were highly motivated to change; it is difficult to imagine that a practice that took the time to sign up to participate in the NDP would not be affected by the highly publicized efforts to transform to a PCMH and thus not change towards the model. Dr. Iliff (6/11/10) asserts, "No wonder practices and patients didn't like it." This assertion is not supported by the evidence. Nowhere in the report or in the NDP evaluation data is there evidence that the practices did not like the NDP model implementation. If fact, many continued to espouse the value of their on-going change (10,11) after the NDP ended. The patients' satisfaction with care DID NOT change over time. The patients' perception of the four pillars of primary care (access, coordination of care, comprehensive care and personal relationship over time) decreased slightly over the 26 months of implementation (12, pS62, table4).

CONTINUED, EVOLVING TRANSLATION OF THE NDP FINDINGS IS NEEDED

While we are delighted that David Margolius (6/19/10), a medical student, describes our findings as "Exciting. Engaging. Transforming." and that representatives from the College of Family Physicians of Canada (Drs. Mang and Gutkin (7/11/10)) describe our findings as a "welcome contribution to the body of international literature that the CFPC has been examining as we further develop a model more appropriate for the Canadian health care system," we agree with Dr. Green's exhortation that a more aggressive dissemination effort is needed. We are mindful for the need to further disseminate our findings (detailed on 2, pS82, table 1) to clinicians, health system administrators, patient advocates, and policy makers so that they can benefit from the experiences shared in this "Voyage of discovery" as described by Dr. Green.

REFERENCES

1. Crabtree BF, Miller WL, McDaniel RR, Stange KC, Nutting PA, Jaen CR. A Survivor's guide for primary care physicians Journal of Family Practice 2009 Aug;58(8):E1-E7.
2. Crabtree, BF, Nutting, PA, Miller, WL, Stange, KC, Stewart EE, Jaen CR. Summary of the National Demonstration Project and Recommendations for the Patient-Centered Medical Home Annals of Family Medicine 2010 Jun;8(Sup 1):S80-S91.
3. Jaen CR, Crabtree, BF, Palmer RF, Ferrer RL, Nutting, PA, Miller, WL, Stewart EE, Wood, R, Davila, M, Stange, KC. Methods for Evaluating Practice Transformation towards a Patient-Centered Medical Home in the National Demonstration Project Annals of Family Medicine 2010 Jun;8(Sup 1):S9-S20.
4. Miller, WL, Crabtree, BF, Nutting, PA, Stange, KC, Jaen CR. Primary Care Practice Development: A Relationship-Centered Approach Annals of Family Medicine 2010 Jun;8(Sup 1):S68-S79.
5. Stange, KC, Nutting, PA, Miller, WL, Jaen CR, Crabtree, BF, Flocke, SA, Gill, JM. Defining and Measuring the Patient-Centered Medical Home Journal of General Internal Medicine 2010 Jun;25(6):601-612.
6. Stange KC, Jaen CR, Flocke SA, Miller WL, Crabtree BF, Zyzanski SJ. The value of a family physician. J Fam Pract 1998 May;46(5):363-368.
7. American Academy of Family Physicians (AAFP), American Academy of Pediatrics (AAP), American College of Physicians (ACP), American Osteopathic Association (AOA). Joint principles of the patient-centered medical home. February 2007. http://www.aafp.org/pcmh/principles.pdf. Accessed September 18, 2010.
8. Patient-Centered Primary Care Collaborative. Web site. http://www.pcpcc.net/. Accessed September 18, 2010.
9. Deloitte Issue Brief: Medical Home 2.0: The Present, the Future. 9/16/2010. http://www.deloitte.com/assets/Dcom-UnitedStates/Local%20Assets/Documents/US_CHS_MedicalHome2_091610.pdf. Accessed September 18, 2010.
10. Philadelphia Inquirer. A new model in medical care. 5/24/2010. http://www.philly.com/inquirer/health_science/daily/20100524_A_new_model_in_medical_care.html. Accessed September 18, 2010.
11. American Medical News. Medical homesteading: Moving forward with Care Coordination. 7/6/2009. http://www.ama-assn.org/amednews/2009/07/06/gvsa0706.htm. Accessed September 18, 2010.
12. Jaen CR, Ferrer RL, Miller, WL, Palmer RF, Wood, R, Davila, M, Stewart EE, Crabtree, BF, Nutting, PA, Stange, KC. Patient Outcomes at 26 months in the Patient-Centered Medical Home National Demonstration Project Annals of Family Medicine 2010 Jun;8(Sup 1):S57-S67.

Competing interests:   None declared

Having anticipated this entire supplement for more than a year, I read all the abstracts immediately and then decided to take my time studying the reports and watch for reactions from individuals and organizations, through the end of June. This past week, I deliberately strolled through the entire supplement, cover to cover, in sequence, and now want to join in the Track discussion (I have also read with great interest all commentary to every article in the supplement posted as of July 1.)

1. This supplement is one of family medicine's all-time bargains, a jewel, and every current and aspiring primary care and family medicine leader would do well to approximately memorize it. We the field got more than we asked for, even if we didn't get from this everything anyone wanted from it. We have been handed an empowering evaluation, much more so than an RCT, an evaluation that was not just an inspection, but a voyage of discovery. We should particularly thank the Guest Editors, Dr. Bayliss and Dr. Phillips for what must have been a challenging task.

2. On my own, I listed 44 insights from the papers, and then was delighted to see most of them captured in Table 1 on page S82. If other readers don't have the inclination to study the whole, I recommend this table and "Primary Care Practice Development: A Relationship-Centered Approach" as the abolute "must read, can't miss" assignment.

3. At this point, it seems to me that key messages from this well- done practice-based research include:

A. The PCMH is useful,even galvanizing, but limited, as a political construct and is best understood as the rallying point for robust, modernized primary care that now necessitates a new mind model from all of medicine, policy makers, and especially those wonderful clinicans and staff members taking care of most of the folks in the US today. PCMH is really not a thing, a collection of techniques; it is presently a journey toward a destination not yet crystal clear. Such is the plight of innovating leaders.

B. Changing the largest platform of formal health care delivery, from a set of concepts from the 19th century using some prudent guesses at the beginning of the 21st century--is HARD! It is not a tweak. It is transformative creation of a fabulous, enabling new platform with rich internal and external relationships, doing care better in ways never before possible. It requires a new "mind model," and new narratives and stories. Three-six year practice development planing is a great piece of advice, and maybe the field can shoot for going over a developmental threshold before 2020.

C. Facilitation works and critical components of it are now known and deliverable. Most of us need some help.

D. Fixing primary care with the ongoing evolution of the PCMH will not be possible as long as the broken US health care system continues, nor until primary care is capitalized. Meanwhile, the clock is counting down on health care that, like a cancer, eats up the nation's gross domestic product at the expense of education, transportation, defense, science, recreation, environment, energy. Perversely, this means getting on with co-evolution of the PCMH, nearly two years after this project's data were collected, is an excellent, timely strategy.

E. Anyone that still thinks primary care is not extremely complex and both intellectually and emotionally challenging is not paying attention or can't read. Anyone aspiring to be the best at personal doctoring and primary care probably must refresh their humility quotient.

F. Move now from front and back offices with an adjudicating overlord called a physician to integrated systems of care with stunning, reliable teamwork, inclusive of mental health and linkage to public health. Do it while the PCMH can evolve. Do it working with others, even if others have not always been a close friend.

I am not at all concerned about all the "hoopla" around patient experience getting worse during a 26 month period and small effect sizes. We would do well to recall that none of these practices had any financial incentives for their efforts. And, what would one expect from a traffic survey of travelers on a road under repair if asked, "Does this ride delight you? Are you getting exactly what you want from this road the way you want it when you want it?" And to lament small effect sizes is akin to lamenting less than gorgeous blossoms on roses planted in a desert, with the gardner having only a cup of water each day.

I wish this supplement had been published sooner, closer to the actual experience. And I am not in complete agreement with all aspects of the reports. Of particular concern is the confounding of personal doctoring and the need for teamwork and a new mind model for practices. The allusion to an intrapractice and extrapractice perspective might point the field toward clarifying the replacement of physician centered practice systems while simultaneously securing the proper role of the personal physician that still appears to be at the center of the relationship that seems necessary for the PCMH to function properly. What, in the future high performance system, will each of us as patients expect and get from "our doctor?" Or is someone proposing that American's will have a payer, a team, a portal, but no one they consider their personal, attending physician?

Of greatest concern to me, at this point, is the apparent lack of agressive, national promotion and dissemination of the results of this stunning piece of work. There is not a peer for this work. It is very special and rich with knowledge that family medicine and primary care need. Where are the celebrations? Where are the rapidly called meetings locally and nationally to digest and debate these results and to make further plans?

Let me conclude with a personal and heartfelt thanks to 36 practices who rose to leadership when their patients, their discipline, and their country needed them.

Competing interests:   I chaired Task Force 1 of the Future of Family Medicine Project that called for a new model of care and an assistance organization, several years ago.

The concept of a medical home is not new, as noted in this set of articles. Although first articulated by the American Academy of Pediatrics in the 1960s, most family physicians at that time were practicing in a way that comes very close to meeting today’s definitions of a patient centered medical home (PCMH). However, the complexity of medicine and society has greatly increased from those days, so the challenges of providing that same type of care have grown exponentially.

This series of papers provide us with the first rigorous and comprehensive evaluation of PCMHs. There is much to rejoice in the findings, as well as reasons for despair. The best news is that a practice can redefine itself in such a way as to not only embrace the principles of a PCMH, but to actually become one. As reported in these papers, many practices were able to implement a significant percentage of the components of a PCMH. Also encouraging are the findings that physician practices can benefit from facilitators. It would be more daunting for most practices to think they would have to walk this path alone.

The not-so-good news is that these achievements came with a high price – they required significant organizational support and resources. Even with those resources, most practices did not implement 100% of the components of a PCMH. Without a major infusion of resources, what proportion of practices in our country will be able to undertake this journey?

Also of interest, patient perception seemed to be negatively impacted in a small way. Why? I have a hypothesis: Having components (even all components) of a medical home in place does not truly make a medical home. A medical home is not one that offers electronic medical records, or that provides patient with a secure web-portal, or even one where the practice can communicate seamlessly with consultants, hospitals, and community resources. A medical home is a place where every single individual in that place lives and breathes the principles of a medical home, and where all of the tools are in place to allow those principles to come play out for every patient in that practice.

As a physician practicing in an NCQA-recognized medical home I can speak from the heart when I say that such a designation is one small step in what will be a long journey for us. Although we have earned the NCQA seal, we haven’t yet done what we are committed to do: to be a true medical home.

Competing interests:   None declared

It's fascinating how people can read the same article and come to very different interpretations.

I see the evaluators' report on the NDP as strongly supporting the concerns that I and many other physicians have had about the PCMH from the start, i.e.:

1. There is no evidence that the seemingly random concatenation of managerial, administrative, and technological chores that make up the PCMH actually do anything to improve patient health and satisfaction; and

2. The PCMH is not financially viable.

Dr. Iliff says correctly that there is no shame in a failed experiment. What is shameful is that the AAFP for many years has been vigorously pushing this model of patient care without any supporting evidence. What is shameful is that the AAFP did not demand that the NDP include a financial analysis of the PCMH. What is shameful is that the AAFP has not been open to discussion on the merits and defects of the PCMH (see Dr. Iliff's "Making It" blog). And what is worse than shameful, what is deeply unethical, is that the AAFP is selling to physicians, through its TransforMED subsidiary, a product that has been shown to be defective, that does not achieve the results claimed for it.

The Orwellian spinning of the evaluators' report has already begun. We're being told that the National DEMONSTRATION PROJECT was never a demonstration project, it was always a "learning laboratory." We're told that anecdotal reports from other practices provide the supporting evidence for the PCMH that did not come out of the NDP. AAFP president Dr. Lori Heim states "the TransforMED NDP achieved what it set out to do." Huh? Did it set out to show that the PCMH doesn't work?

It will be very interesting to see how the AAFP responds to this report. Will they continue to cling to a misbegotten model of patient care, or will they decide to truly represent the interests of America's family physicians?

Competing interests:   None declared

Exciting. Engaging. Transforming.

These are the words that replace “overwhelming”, “isolated”, and “dying”. The future of primary care is bright, and I, as a medical student, can’t wait to be a part of a transformation similar to the one implemented by TransforMED in the National Demonstration Project (NDP).

The primary care physician workforce is in a death spiral. In the United States, the need for quality primary care is greater than ever. Primary care needs more than incremental steps, it needs a leap(1).

As the NDP evaluation team points out in their summary paper, the NDP effort to tranform primary care was not tranformative enough(2). Too many of the components of the NDP simply added work to an already overburdened primary care physician. The next try has to go even further: team- based care, more alternative encounter types, and eventually payment reform(3). That is the primary care that I want to be a part of; the work of TranforMED indicates we are headed in the right direction. And as the words “exciting”, “engaging”, and “transforming” become inextricably linked to primary care, I believe many more medical students will follow. This is a victory of the NDP worth noting.

(1) Bodenheimer, T. (2006). "Primary care--will it survive?" N Engl J Med 355(9): 861-864.

(2) Crabtree, B. F., P. A. Nutting, et al. (2010). "Summary of the National Demonstration Project and recommendations for the patient-centered medical home." Ann Fam Med 8 Suppl 1: S80-90.

(3) Margolius, D. and T. Bodenheimer (2010). "Transforming primary care: from past practice to the practice of the future." Health Aff (Millwood) 29(5): 779-784.

Competing interests:   None declared

The health care community owes a great debt of gratitude to the designers and participants in the NDP. The rich qualitative and quantitative results outlined in these eight articles mark a significant movement forward in our understanding of the work that will be necessary if we are to revitalize primary care in the U.S.

Many tools outlined in the project will useful to those of us working in practice transformation: Supplemental appendix 4 Clinician Staff Questionaire; Supplemental appendix 10 TransforMED Implementation Index; Six step method for resolving conflict; The Key to implementing Change in your Practice; Table 2. Components of the Patient-Rated PCMH Scale.

The importance of building adaptive reserve by means of effective team meetings strongly resonates with the experience of our own practice over the past few years.

It was especially interesting to note that some of the newer concepts promoted for enhancing the medical home may require more research, i.e. group visits (9 of 15 practices discontinued group visits) and e-visits (which appear to have been trumped by same day access.)

I would propose two challenges to the TransforMED working group: 1.) Can we design more robust tools for measuring the “four pillars” of primary care (which could then be incorporated into the NCQA Medical Home criteria)? 2.) How can the patient response rate to surveys be improved? (The response rates listed of 21%-29% would seem to limit the strength of the outcomes data.)

Again, thank you for this perceptive and pioneering study.

Competing interests:   None declared

I am pleased to read the publication of the NDP results to date.  Because primary care is the foundation of high performing health systems and because primary care is suffering in delivery, it makes very good sense to test approaches to better deliver primary care.  

The hypothesis the NDP tested centered on asking practices to adopt a laundry list of 'changes' to improve patient care, and to see if a consultancy model improved the adoption of those changes.  We see from the evaluation teams a disappointing set of results:  some medical chart markers of chronic conditions have improved for some patients but overall patient experience of care worsened.  This is very unfortunate as overall patient experience of care better predicts overall outcomes for a population than tracking a handful of chart markers for a handful of patients. 

We now have a much better idea that this laundry list of changes and consultancy approach does not lead to the outcomes our country needs: improved POPULATION health (as opposed to a handful of improvement for a handful of conditions), and improved TOTAL cost of care.  This is very important information because the NDP was a fore-runner to an entire parade of projects built on the same hypothesis.  

Across the nation there are projects in the works matching the NDP strategy:  practices are asked to work through a similar laundry list of 'changes' to improve a handful of chart indicators for a handful of patients with the expectation that this will lead to improved POPULATION health and total cost of care.

If the fore-runner's results are minimal to negative we ought to be very concerned by the uniform, lock-step homogeneity of projects in the works.  This national focus on one (now questionable) solution is risky, unwarranted, and ultimately unnecessary.  

Running at the same time as the AAFP's NDP, the Ideal Medical Practices project demonstrated population health improvement, improved experience of care, reduced hospitalization, reduced emergency room use, and improved metrics of chronic conditions.*  We did not have to trade one thing for another.  

We all deserve a better chance at our real goals than a single-minded approach that has so far achieved little that patients and the public would find meaningful. If our nation is really interested in supporting better care to achieve better outcomes and bend the cost curve it ought to investigate and invest in projects better aligned with transformation as opposed to marginal improvement.

*Wasson JH, Anders SG, Moore LG, Ho L, et al. Clinical Mircosystems, Part 2: Learning from Micro Practices About Providing Patients the Care they Want and Need. Joint Commission Journal on Quality and Patient Safety, August 2008, 34(8) pp. 445-452.

Competing interests:   President, Ideal Medical Practices a non profit corporation supporting those who would deliver superb care to their patients