If one could name a radiologist who one might describe as being seemingly virtually everywhere at once, one would have to put Keith J. Dreyer, M.D. at the top of any list for consideration. Dr. Dreyer’s core appointments are as the vice chairman of radiology informatics at Massachusetts General Hospital in Boston, and an assistant professor of radiology at Harvard Medical School. Not only does Dreyer hold those faculty and organizational positions; he has also served as co-chair of the informatics committee at the Reston, Va.-based American College of Radiology (ACR), and as chairman of that association’s government relations informatics committee.

 In fact, it was through his involvement with informatics issues at the ACR that Dreyer became more and more aware of the need to create a nationwide set of services to provide better clinical decision support at the point of care for practicing radiologists.
The efforts of Dreyer and a number of other colleagues led to a pilot project sponsored by the ACR that ultimately led to the commercialization of a solution through ACRselect/the National Decision Support Company, led by Michael Mardini, founder and CEO, and which has signed contracts with a number of major integrated health systems in the U.S., several of which are going live with that decision support solution in early 2013.

Dreyer sat down with HCI Editor-in-Chief Mark Hagland during RSNA 2012, the Radiological Society of North America’s annual conference, being held at the McCormick Place Convention Center in Chicago Nov. 25-29, to talk about clinical decision support for radiologists, and to discuss more broadly the challenges and opportunities facing radiology in the emerging healthcare environment. Below are excerpts from that interview.

Tell me about your involvement in what has become ACR Select/the National Decision Support Company.

One of my main motivations for becoming involved in the ACR was getting involved in what has become that entity. We had demonstrated the value of doing this—providing clinical decision support at the point of care—at Mass General, but its value was restricted to our organization. What’s more, it was difficult to prove the value of clinical decision support with such a limited scope. A key moment for me in all this was a meeting that I had in 2009 with Keith Faulkner, who was the healthcare oversight for the OMB [the White House’s Office of Management and Budget] at the time. He asked me pointedly what the difference was between decision support and radiology benefit management (RBM); he couldn’t discern the practical difference between the two concepts. And I came away from that meeting, and stepped back and said he was right. And it was important to create a nationwide entity that could provide robust clinical decision support that would provide radiologists with a better solution than what they got out of RBM, which many radiologists find oppressive.

So I got involved in taking this concept national. Henry Ford Hospital and the University of Wisconsin’s 16 hospitals became beta testers beginning in July 2011, and the commercialization came out in July, with Mike Mardini. The official first release of the solution came on Nov. 1, and the implementations of the product will begin after the first of the year.

Keith J. Dreyer, M.D.

Where are radiologists in terms of understanding how automation-facilitated they will be, and how much they will be a part of the new accountability and transparency in healthcare?

The first step of accountable care, in terms of demonstrating your metrics, is meaningful use. And radiologists are really trying to comply with meaningful use. In a lot of cases, most of the meaningful use requirements—and I think they were written very elegantly—were designed for office-based physicians in private practice, for primary care physicians. They were really not designed for specialists. But radiologists were there in the imaging center setting—they can control their space and decided whether they want to buy certified products like PACS, RIS, and EHRs. But in the scenario where the radiology group is affiliated with a hospital or multiple hospitals, it’s up to the hospital to decide what technology to buy. So CMS came back and said, we’ll provide an exception to radiologists, anesthesiologists, and pathologists, that for those essentially hospital-based specialists, we’ll create a hardship exemption to not participate but also not receive benefits, or penalties, for five years.

And they still want to participate. So back to your original question: I think like everybody else, they’re mindful of what’s happening at the federal level; and they want to participate in meaningful use, not only because of the incentives, but also to participate in quality. And what is it that they should be doing to participate, as the fee-for-service model gets shifted? So they’re keen to be a part of meaningful use. And that exposes them to all these clinical quality measures (CQMs). So it’s like a chicken-or-egg situation. CMS [the federal Centers for Medicare & Medicaid Services] creates these programs for physicians generally, and then the specialists say, we want to participate. So then CMS modifies those programs. And that’s a step-wise progression towards accountable care.

Is the culture of specialists now changing? Are radiologists coming to accept the new parameters around medical practice?

Maybe I’m being mildly defensive as a physician. Here’s what I’ll say: I have a resistance to change across my whole life—don’t make me change to a Mac, etc. So there’s general resistance among all people to change. But in medicine, partly because of fear of litigation and everything else, you don’t want to be the first bird, nor the last; you want to be in the middle. There’s that comfort with being in the middle of the group. And the feds want to push change. And that’s what I think is happening here; no one’s scared to death to do this. People are just saying, tell me what I have to do. Because in the early stage of decision support, 2004, 2005, all you had to do was to talk to a physician who had used RBM who had been told ‘no,’ and all you had to do was to show them decision support, and they were there. Because with the RBM process, you’ve wasted a lot of time, and either you end up doing what you wanted to do in the first place, or you can’t now, and you can’t do what you wanted to do that you thought was best for the patient.

And here, with decision support, the doctors are really learning something. So it really is the perfect situation for how we want to practice. And the other thing I hear is, patients are getting so smart, and so specific. Patients will come in and say, I want an MR on my knee. And now with this, the physician will say, actually, according to national standards, that’s not called for. And you couldn’t do that with an RBM.

So more and more, physicians will be rationalized into and accept, parameters?

Yes, because payment will be put into parameters based on quality standards. It won’t be like the days of HMOs where they said, here’s $3,000, now work with that. And also, patients will be faced with a choice, and they could choose treatment outside your organization. So you really have to become patient-aware, consumer-aware. And if a patient has to sit there for three hours or has a bad experience with a doctor, they’ll move.

How will radiology practice in the next three years?

I gave a plenary oration here at RSNA, and I was talking about meaningful use, but in a broader context. And I said, we got ourselves in the state we’re in because of a fee-for-service payment model. So we’ve been driven by quantity. But it made us less relevant and less present in patient care; it all becomes about being revenue-conscious and maximizing volume. And if you squeeze fee-for-service down so much that you just can’t do it more and it becomes a commodity, I think the system will shift towards quality and being present for patients and patient-centered. And in terms of teleradiology, we didn’t give images to our peers nearby or to our patients; but it’s going to change so we’ll be incentivized to share with our colleagues and competitors, and with our patients, share images with all of them. And technology is changing all that. And so all of a sudden, I’m sharing images. And this technology-driven wave will force change.

So what I argue to radiologists is, we need to get ahead of this wave. And if I have to talk to referring physicians and to patients and that slows me down, then we should be compensated for that patient interaction and communication. So I argue, shouldn’t we be renegotiating our contracts? And some colleagues are figuring that out and some aren’t. and the pioneers will be reengaging radiologists on quality and service, instead of just cranking out interpretations. And these innovators are going to figure that out first.

Instead of becoming hamsters running faster and faster on wheels, right?

I say, imagine you’re talking to a major PACS vendor. And they ask the radiologists, what do you want to do if someone is an ordering physician, and they’re spending a long time looking at a diagnostic image—and for me, the radiologist, what would I want to have happen? What would we want for that ordering physician? If I was volume-based, I’d probably want them to order another exam; that would be how I would design the software. But instead, if I was motivated by outcomes, and I was getting a negative for the exams being ordered, first, I’d be putting decision support in front; and if I’m driven by outcomes, if that patient does well and lives longer, I’d facilitate the ordering physician being able to instant message me, to get additional consultation, as opposed to just ordering more exams. And thirdly, if I’m motivated to provide more patient access, wouldn’t I want the patient to see the images, too? To facilitate the patient viewing images and talking with me—both kinds of interaction. Patients talk to all other physicians, after all. We always hide behind this cloak of, well, they’re really the referring physicians’ patients, we shouldn’t step on those doctors’ toes. But honestly, the patients are fearful of what’s happening, they want to be reassured, and now we can do it in that new model. And the features of the technology will be driven by everybody’s incentives.