The Patient-Centered Medical Home in Diabetes Care

Dennis P. Scanlon, PhD: We’ve all heard the adage that even non-nuanced scientific findings take 18 years to get put into practice; but when we have conflicting results or results that may vary, pending patient characteristics or patient populations, I imagine that becomes much more confusing.

How do we help clinicians in the clinics and in the practices navigate this scientific evidence as it comes out, and help to interpret it so that they’re not spending their days reading? (They’re treating patients, not reading JAMA or the other journals.)

Robert Gabbay, MD, PhD, FACP: I think that’s where professional societies, or specialists, have a role to be able to sift through the data and interpret it. I think the other area where we can make more strides is with decision-support tools for providers at the point of care that would say, “Based on these characteristics, the patient you’re seeing fits into a category that’s very similar to the SPRINT study group or another study group. Here’s what would be recommended.” I think there’s a little of that [happening], and I think you’re going to see more of that in the near future.

John A. Johnson, MD, MBA: That’s actually what happens on the managed care side. We publish what are called, clinical practice guidelines, and we make those available as tools to the clinicians in our network as a guide to help drive some of their decisions— giving them some autonomy to individualize the care plan because again, as we found out, one solution doesn’t solve it for each diabetic.

Zachary T. Bloomgarden, MD, MACE: But, we have to make sure that those guidelines are not 18 years old.

Robert Gabbay, MD, PhD, FACP: Absolutely right.

Zachary T. Bloomgarden, MD, MACE: That’s the dilemma. I’m still getting emails or letters from various pharmacy benefit managers saying, “Here’s a patient with diabetes who is not receiving an ACE (angiotensin-converting-enzyme) inhibitor or an ARB (angiotensin II receptor blocker).” Of course they don’t know that the patient has had hyperkalemia and was intolerant of those medicines, or has low blood pressure.

We really need to somehow integrate the journal findings and the latest developments into the latest understanding of professional societies for [tools that] we provide to clinicians to help them manage their patients.

Robert Gabbay, MD, PhD, FACP: I think the data needs to be a lot better. I think that’s one of the challenges. Each group that is trying to track this has their own sets of data. We, as clinicians, have the data in our electronic health records. From a payer perspective, you have your data, which is not necessarily the same. So, integrating that more effectively would be a huge step forward.

Dennis P. Scanlon, PhD: That’s happened in a lot of areas of healthcare. Dartmouth has obviously made the study of variation across clinical practice quite well-known and a lot could be learned there. There’s clearly variation here, but some of that variation is warranted based on clinical characteristics and patient comorbidities.

I guess the question is, how do we address that recognized need for customization, but bring as much evidence to [the table]?

John, I wanted to go back to you. It sounds like you’re trying to do some things as a health plan to educate providers. How receptive are they? The health plan is coming at you with a clinical practice guideline. I could imagine that being received well in some cases, and perhaps not so well in others.

John A. Johnson, MD, MBA: You nailed it. There are some providers that don’t mind partnering with the health plan, and then there are other providers that feel like, “I’ve been trained this way in residency, this is the latest my medical society is telling me, and I’m going to follow that guideline.”

Again, we just like to stay on the sideline as a partner to ensure that whatever method they’re using is driving toward quality improvement. Whatever path gets them there in the safest tolerable way for our members is really the ultimate goal.

Dennis P. Scanlon, PhD: Bob, you talked about data and the opportunity to assemble better data. What might that look like? I could imagine you would have characteristics of patients (maybe some of their history), different drug regimens and choices, and information on behaviors, exercise, diets, and other clinical comorbidities.

You can imagine putting that together in a data set and looking at associations—this would not be done with experimental studies, but I could imagine some of that could be valuable from a clinical perspective.

Robert Gabbay, MD, PhD, FACP: Absolutely. But even with the data that we already have available, I think we can have better tools. So, for example, at Joslin we’ve developed a Joslin Clinical Analytical Tool that tries to take data that is in the electronic health record, that all providers have—medications, laboratory data, some basic demographics—in an attempt to create different goals for the population and subgroups.

I think the field is at its infancy in terms of doing that, but that’s what we need. We need more nuanced measures of quality that take into account a lot of other factors, and we’re struggling. Patient-reported information is rarely ever integrated into any of these kind of measures—much less socioeconomic status. You could imagine a whole series of variables that could be included, even in the first step.

We found this [to be] a helpful tool in working with clinicians around the country to guide them on subpopulations where there are gaps of care that could be addressed.

Michael Gardner, MD: The AAMC (Association of American Medical Colleges) recently sponsored a project to bring together numerous hospitals to contribute data in a de-identified form for research so that you can look at a much larger population than what is in any single health record system. I think we need more of that kind of collaboration between hospitals—private, public, universities, large healthcare systems—to bring the data together so that we can look at the entire population.

Robert Gabbay, MD, PhD, FACP: One thing, though. I don’t want the message to be that that’s what we need in order to improve care. Right now, you can identify—and I don’t think this happens in a lot of primary care practices—patients that have an A1C (glycated hemoglobin) greater than 9 that have not been seen in 6 months and do outreach. That’s clearly effective and that doesn’t need to be complicated.

Michael Gardner, MD: Certainly, you don’t want to forget the micro level when trying to address the macro.

John A. Johnson, MD, MBA: That’s also under the belief that the provider has an electronic medical record system.

John A. Johnson, MD, MBA: And is able to effectively do population health, which gets to the whole patient-centered medical home (PCMH) model, that a lot of the managed care community endorses—the accountable care organizations and PCMHs—to incentivize providers to share in the risk of managing patients to improve outcomes. So, a better way to do that would be with an electronic medical record.

Robert Gabbay, MD, PhD, FACP: Sure. And fortunately, I think we’ve made great strides in the US in terms of the number of providers that have electronic health records. It’s still not where it needs to be, and then there’s the challenge that those electronic health records are not always great systems to be able to extract data from—that’s obviously the big challenge.

John A. Johnson, MD, MBA: So many different platforms that don’t speak to each other.

Robert Gabbay, MD, PhD, FACP: Yeah, right.

Dennis P. Scanlon, PhD: You mentioned the PCMH model. How is that playing out for diabetic patients? There’s an element of having a coordinated focus of care, but there’s also, as we’ve talked about, the need for a care team—educators, nutritionists, potentially, pharmacist involvement.

After-hours appointments or being available on weekends—some of those components of the PCMH are good. But there are other dimensions, as well, when we think about care for diabetes. You’ve studied this, Bob.

Robert Gabbay, MD, PhD, FACP: A lot of the early pilots around the patient-centered medical home really revolved around diabetes for the reason that a lot of healthcare changes have always been around diabetes. It’s a common disease, it’s measurable in large part—although you might debate where the goals should be—and it’s very costly. It’s a natural target.

If you look at all the evidence, there have been significant improvements in diabetes care by moving towards a patient-centered medical home model. But, that is not the full answer because, again, only a portion of healthcare happens within that practice. Small practices may not have all of the different types of professionals that can help.

You mentioned diabetes educators and dieticians, but the typical primary care practice isn’t going to have that embedded in the practice. That’s where centers of excellence can, in diabetes, be able to send people—educators, dieticians—to various primary care practices once a week or whatever.