Trends in Quality During Medical Home Transformation

Thank you for your typically insightful comment, Dr. Iliff.

Though its representatives, surprisingly, claim "no conflicting interests," the AAFP is in the business of selling the PCMH to physicians, hospitals, and insurers. When they made the decision to do so, they gave up any claim to objectivity regarding the strengths and weaknesses of their very problematic product.

Competing interests:   None declared

A front page article in yesterday's Wall Street Journal reports that scientists at Bayer A.G. failed to replicate results of studies published in respected peer-reviewed journals 64% of the time; only 21% were fully replicated. The pharmaceutical researchers had every incentive to see positive results, but also a counterbalancing incentive to not waste corporate time and resources on wild-goose chases.

This is further evidence of an long-acknowledged problem: that "positive" results are far more likely to be published, while "negative" results are flushed. In fact, this seems to be exactly the desires of some respondents to this article.

Therefore I find it highly commendable that the Annals editors had the wisdom and courage to publish this article, as they did previous cautionary reviews of the initial TransforMed experience. Thanks for taking the heat, folks-- you're true scientists.

On the other side we have the propagandists, with various self- interested motives for suppressing less than laudatory results. That's the problem. That's why we don't get the whole story, scientifically, and waste an enormous amount of money being whiplashed by conflicting practice recommendations.

"Underpromising and overdelivering," as recommended by Dr. Solberg, should indeed be the desire of every researcher and editor.

Competing interests:   None declared

Dr Leif Solberg thank you for your comments and the heading "PCMH is a good thing" -- my concern is not what was said and not because the "data -based confirmation of improvement with medical home changes was not as dramatic as they would like to have seen" , rather but how it was said.

The problem is that is frankly unlikely that policy-makers look beyond the conclusion in the abstract. Ask yourself would your member Congressman look beyond the conclusions in the abstract. I do not think so. !! Rather and simply they will have only the headlines pointed out to them by someone who is not trying to help you Primary care but in fact has economic reason to harm. It is my concern that the way this is written as this article does in fact requires you to look past the headlines to reach the right conclusion about what the paper really has to say. When an abstract leaves one with a conclusion that is not what is in the body of the paper and structured in a way that it can and already has be use to say see "SEE we told you need to stick to disintegrated uncoordinated episodic care system" than quote your abstract - that worries me and it has already happened. I would guess that was not how you wanted your abstract used and certainly not what your paper is saying as I read it. We live in a world of sound bits and they matter.

Competing interests:   None declared

It would seem that accredited Medical Homes have done a good job building the infrastructure, e.g., EMR/Registries, care coordination and team care, necessary for PCMH 1.0. A critical missing element in PCMH 1.0 is a total lack of focus on patient-centered communications - a key driver of patient-centered outcomes. In other words, there is no evidence to suggest that physicians in accredited Patient-Centered Medical Homes are any more inclined to use patient-centered communication techniques, i.e., Four Habit Mode, than any non-PCMH practice. Until we change the way physicians and patients relate and interact with one another, it is unreasonable to expect the kinds of "hoped for" changes expected from PCMH. PCMH 2.0 must focus more on such change.

Competing interests:   Consultant and Solution Vendor

Ms. Coleman, similar to others' comments, takes exception to our use of NCQA certification as our marker for being a Medical Home. During the years encompassed by this study (2005-2009), the PPC-PCMH was one of the few means by which practices that embraced the tenets of primary care transformation could achieve recognition. For practices not participating in a demonstration project such as the one AAFP sponsored with TransforMED, the NCQA application process provided a tangible structure to focus efforts for practice improvement, particularly in the areas of access, care coordination, and panel management through registries. Was it the ideal? No. The 2008 recognition standards were widely acknowledged to be lacking in dimensions of patient-centeredness and comprehensiveness.1 Indeed, the NCQA itself responded to such criticisms in developing the 2011 standards, which Dr. McGeeny allows are a "definite improvement."

The more relevant question is how our choice of the NCQA marker reflects on our conclusions. Coleman seems to be arguing that the HPMG clinics were not really "transforming" during the study period and that's why larger improvements weren't seen. Dr. Solberg has already replied that data for the period prior to our analysis (2000-2005) showed the rate of improvement in HP clinics was similar then. Specifically, on the Consumer Choice patient survey, the improvements did not exceed 1-1.5% per year.

It's also worth noting that in 2008 the Minnesota state legislature mandated the creation of Health Care Homes (HCHs), defined through a multi -stakeholder process that included patient advocacy groups, health plans, and medical professional societies, among others. The criteria for Minnesota HCH certification attempt to incorporate such patient-centered initiatives as care coordination and care plans with patient input at team meetings. Having represented Minnesota family physicians as president of the MAFP during key months of the process, I will be among the first to admit that it is in its infancy and that filling out yet another application, and even passing the required site visit, is no guarantee that fully meaningful PCMH "transformation" is in place within a particular practice. Nevertheless, the 21 HP medical practices were again within the group of early adopters to obtain Minnesota HCH certification in 2010. So a reasonable conclusion would be that even in the "advanced health care market" in our state, practices that are involved in ongoing efforts for continuous quality improvement and are in a position to incorporate innovative methods of care delivery exemplified by the PCMH, will continue to make measureable improvements according to the validated quality and satisfaction measures we studied.

As authors we attempted to be candid about limitations to generalizability of our findings. However, the PCMH will likely be applied in health care markets of all sorts, in varying stages of development. We remain convinced of the validity of conclusions and appreciate the spirited discussion the work engendered.

1. Malouin RA, Starfield S, Sepulveda MJ. Evaluating the tools used to assess the Medical Home. Managed Care,June 2009;44-48.

Competing interests:   None declared

I have read this article by Solberg et al with great interest. While the concept of patient-centered medical home (PCMH) has gained much attention recently, there has been relatively little documented evidence that PCMH can fulfill its lofty promise of achieving the "triple aim." I suspect there are two reasons for this: First, many of the existing PCMH implementations are still in infancy and evolving, so there is relatively little data to show their full effects. Second, implementations of PCMH are driven and shaped largely by the sponsoring organizations' unique needs and goals, rather than by any desire for scientifically rigorous research design. This is not necessarily a criticism but rather a reality that a researcher like myself must face in evaluating the effects of PCMH.

This research by Solberg et al is unique in that the authors do have substantial data over many years, but their ability to make generalizable inferences is limited because of a lack of data from a plausible comparison group. The glaring weakness in their research design is of course the lack of Picker survey results from non-HPMG patients that could serve as a direct benchmark. I applaud the authors for attempting to address this problem in a different way, and I do think that the results they present are encouraging, albeit inconclusive, evidence in favor of PCMH.

Geisinger Health System has implemented its version of PCMH (referred to as ProvenHealth Navigator?, or PHN) since November of 2006. My colleagues and I at Geisinger have recently published an article that shows that PHN is associated with improvements in certain patient outcomes (Maeng DD, Graf TR, Davis DE, Tomcavage J, Bloom FJ Jr. Can a Patient- Centered Medical Home Lead to Better Patient Outcomes? The Quality Implications of Geisinger's ProvenHealth Navigator. Am J Med Qual. 2011 Aug 18. [Epub ahead of print]). Currently, we are also analyzing our own data to show that PHN is associated with significant reductions in cost and improvements in patient satisfaction. Our findings from these studies should nicely complement existing knowledge about the effect of PCMH.

For the two reasons I mentioned above, it is extremely difficult to conduct a single study that proves the effectiveness of PHN. However, as more evidence accumulates from additional future studies, I believe a strong case can be made in favor of PCMH as the way to reshape the healthcare system.

Competing interests:   None declared

I agree with many of the previous comments and won't repeat their good points here. But I would ask this: On page 516, the authors say "except for implenting advanced access scheduling in 2000 and the EHR in 2002-2004, there have been no abrupt changes in approach to care worth noting..." Why then would we expect that there would be abrupt improvements in the HPMG sites as compared to their community peers? Said another way, if there was no transformation in care why would we expect transformation in patient satisfaction and health outcomes? The medical home movement must get beyond looking at the completion of an NCQA application as "transformation" and get real about making changes to the way care is delivered if we want to see improvements in patient experience, outcomes and cost.

Competing interests:   None declared

Seeing the reactions to our study of medical home transformation in Minnesota has been a pretty interesting experience. On the one hand, it is gratifying to get any reaction to a published research study, especially when that reaction comes from leaders like Paul Grundy, Roland Goertz, and Terry McGeeney. On the other hand, it is troubling that because our data-based confirmation of improvement with medical home changes was not as dramatic as they would like to have seen, they felt the need to attack and dismiss those data and our conclusions. It is also troubling to be accused of harming the cause of medical homes, when I am one of the greatest boosters of the kind of changes represented by the PCMH. I just think underpromising and overdelivering is a better strategy in the long run than the opposite - the quickest route to damaging PCMH sustainability is to create lofty expectations that can't be achieved.

I especially would have thought that Dr. McGeeney might have been more cautious than to say that "two to three years for transformation is generally adequate." There wasn't much I saw in the thorough evaluation of the TransforMED experience to support that expectation. After 26 months of assistance, the intervention clinics improved in their quality scores by just about the same amount as our experience (2.5-4%/year vs. our study's 2-7%/year) while there were no improvements in patient-rated outcomes while our group improved at 1-3%/year.

Similarly, Dr. Goertz disagrees with our findings, citing changes in the BCBS of ND initiative that saw "tremendous improvements over a three- year period." However, the cited improvements in blood pressure and LDL control for diabetes were only 2%/year and the idea that one can compare emergency room visits and hospital admission rates in clinics that volunteered to participate in this initiative with those that didn't speaks for itself. He also says that the hardest work in our medical group had been done before the study began. Although there wasn't room in the article to provide data for the period prior to when we had comparable data between HPMG and other local groups, the rate of improvement was similar then.

These important PCMH leaders seem to feel that anything but wild enthusiasm is harmful to primary care and its current redesign, and that "the data and evidence are conclusive." As a strong supporter of the PCMH who thought that the data in the article were overall helpful to the cause, and as a student of the research literature on the topic, I regretfully will have to disagree.

Competing interests:   None declared

As many of you know, I represent a very large employer that spends a staggering amount of money each year on health care for employees. During the past 20 years, we have seen a wide scale, rapid decline in quality compared to other developed nations. In the past year alone, three papers published in the journal Health Affairs show us that the United States continues to decline in value, while having the highest rate of increase in costs. I have no doubt that the problems lie within the current delivery system. That said, I welcome slow, incremental improvement over the status quo.

I do not get the point of this paper. It seems to me the authors are comparing newly minted NCQA PCMHs to a community where many practices in the state have been migrating toward the Minnesota Patient-Centered Health Home (MPCHH) model for a long period of time. From my perspective, this article is saying that with transformation to NCQA PCMH, you get only small improvements, but when you transform to a MPCHH over many years, it is a success. It begs the question: why in North Dakota and nearby Michigan do we see large, rapid. and positive PCMH transformation? (Patient-Centered Medical Homes In Michigan Thomas Simmer Health Aff November 2011 30:2217; doi:10.1377/hlthaff.2011.1102 http://pcpcc.net/files/102011summit/5.hanekom_pcpcc_10-21-2011.pdf)

The number one finding of a recent Commonwealth Fund study is that the United States spends more on health care than other countries, but it doesn't get the most out of the health care system," said Karen Davis, president of The Commonwealth Fund. "The United States performs the most poorly on access to care and the financial burden that comes with chronic illness." The study also found that people who have a patient-centered medical home -- a primary care physician or practice that coordinates treatment across specialties -- felt better about their care and were less likely to report medical errors. http://content.healthaffairs.org/content/early/2011/11/02/hlthaff.2011.0923

I read Dr. Solberg's study three times and am still scratching my head trying to figure out why Annals of Family Medicine would publish it. I am sure Solberg's assertion that "expectations for large and rapid change are probably unrealistic" will probably be used against me as I continue my quest to purchase PCMH-level care for my employees and their families. I suspect, but hope I am wrong, that this article will do more harm than good toward getting primary care away from buying episodic, uncoordinated, inaccessible, and dis-integrated care.

Nearly all primary health care stakeholders want to learn from successful PCMH initiatives. That is why it is so important that we in primary care unite and collaborate to make the PCMH model the strongest it can be - not tear it down. Nothing positive will come from "throwing the baby out with the bath water." We have enough key learnings at this point to successfully build upon the work done by our PCMH pioneers and take it to the next level. This is critical to the health of our patients and the health of our "health care system."

Competing interests:   None declared

One key learning over time is that all of the attributes of the PCMH model of care are interdependent. There is no data to support that leveraging just some of the attributes of the PCMH leads to incremental improvement in efficiency or quality. In fact, the full value of the PCMH model may not be recognized until a practice is a complete PCMH, which current recognition tools cannot assess or validate. While the HPMG practices should be applauded for their 11 year journey of practice transformation, it is important to note that such a length of time is not required for practice transformation. The TransforMED experience of two to three years for transformation is generally adequate with the right support and incentives.

NCQA recognition is often used, as it is in this article, to validate a practice as a PCMH. While the 2011 NCQA standards are a definite improvement, the 2008 standards were lacking to the point of creating more confusion than added value. 2008 NCQA standards were woefully inadequate in assessing patient engagement. Patient satisfaction surveys do not necessarily engage patients in their care. The 2008 standards also did very little to recognize care management and care coordination in a practice. Care continuity requirements in the standards could be met without following up on emergency room or hospital discharges. The same is true for consistent follow up on lab tests and referrals.

The access-to-care criteria of the 2008 standards were more about triage than actual practice, including weekend and evening hours as well as same-day appointments. While population management is critical to the future of primary care and the PCMH model, passing only one "process" standard was required for recognition. Most everyone agrees that team care is essential in the PCMH, yet NCQA team requirements are minimal. Dr. Solberg noted in the article that nurse practitioners and physician assistants are part of the care team at HPMG, but NCQA recognition in no way equates to effective team care. Finally, the lack of recognition for full "comprehensive care" in the recognition standards for primary care practices is significant. This lapse is potentially amplified in a large multi-specialty group, as I witnessed during my many years of practice experience.

Other attributes of the PCMH model that are important for its success and not generally valued by current recognition entities include: active efforts to engage the patient in every care opportunity; designated care managers in practices; functioning, self-populating, searchable data bases to be leveraged by care managers; payment incentives based on care efficiency and quality by the providers; patients' access to their data and care team via an EMR portal; and risk stratification of the patient population.

I applaud Dr Solberg's efforts in this article to assess the impact of the PCMH model in a high-functioning multi-specialty group. The article does a great job of demonstrating the challenges posed during the PCMH evaluation process. Until there is a valid, reliable methodology to ensure that all of the interdependent attributes of the PCMH model are being leveraged consistently, we will continue to struggle with anecdotal experiences from projects and pilots. That is not to say that the PCMH model does not hold great promise for the health care system. Indeed, the health care system will most likely transform according to the attributes of the PCMH long before anyone figures out how to study it. The urgency and need is too great, and we know enough to forge ahead.

Competing interests:   None declared