Radiology’s Value Proposition: The Massachusetts Experience

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

On September 9, 2013, Cheryl Proval, editor, Radiology Business Journal, moderated “The Boston Experience: Radiology’s Value Proposition in the New Health-care Paradigm,” a panel discussion held in Boston, Massachusetts, at the RBMA Fall Educational Conference.  The meeting location offered a unique opportunity to explore radiology’s changing value proposition under health-care reform with a panel that represents hospital-based radiology, outpatient radiology, and the payor community.

Steven E. Seltzer, MD, FACR, has been chair of the department of radiology at Brigham and Women’s Hospital (Boston, Massachusetts) and Philip H. Cook professor of radiology at Harvard Medical School since 1997.

Michael Sherman, MD, MBA, is senior vice president and CMO of Harvard Pilgrim Health Care (Wellesley, Massachusetts), a full-service health-benefits company. He serves on the advisory board of the Institute for Clinical and Economic Review and on the board of directors of Massachusetts Health Quality Partners.

Patricia Whelan-Downey, MHA, is CIO of Shields Health Care Group (Quincy, Massachusetts), a leading provider of outpatient diagnostic imaging for 25 years, and is senior partner of Shields Consulting.


Proval: Massachusetts is the birthplace not only of the American Revolution, but also of the revolution in US health care. The state went live, in 2007, with its brand of health-care reform, widely seen as the model for the federal Patient Protection and Affordable Care Act.  How did health-care reform affect what Harvard Pilgrim was looking for from radiology?

Sherman: In Massachusetts, we were first—ahead of the curve—in universal access. I think we are getting a lot of scrutiny to answer these questions: What happens after you do that, and what happens to cost? The good news is that virtually all of our citizens in Massachusetts have coverage.

We are number one in many things, however, and unfortunately, one of those things is per-capita cost (among the 50 states) in what we spend on health care for our population. Part of it is that we have world-class providers, and some of them are expensive. It’s an expensive part of the country, too.

We also have a state that likes mandates. We have mandates that many other states do not have concerning autism, infertility, and other conditions. These are not inherently good or bad, but when you put those in place, they do raise costs.

Jump ahead from the onset of universal access to today: We’re in an expensive market, and there’s concern about our ability to pay for care. Massachusetts Chapter 224 passed last year; that is a law that says, OK, good job getting access: Now, we need to think about value.

One of its implications is that the state has decided that health-care costs should not grow any faster than 3.6% per year (which policymakers estimated as the probable equivalent of growth in the state’s version of the gross domestic product). This has payors and providers thinking, together, about how to work toward a 3.6% goal.

It’s having an impact on how we contract—and on how we look at coverage policies (in terms of products, narrow networks, lower price points, and other types of innovative products that we have. It has to do with how we are looking at our informatics capabilities and how we are partnering with the physician community, which presumably would have some enforcement mandate.

That’s important stuff, but ultimately, we’re not quite sure what happens if we don’t keep cost increases at less than 3.6%. It’s been suggested that you’ll be taken before the state attorney general’s office.

We did partner with a radiology benefit management (RBM) company, and that’s helped keep our trend to 3.6% for radiology, over the past five years. It was higher, before that.

Proval: What did Brigham and Women’s Hospital want from you?

Seltzer: We are in the sixth year of what we jokingly call RomneyCare (though it was disavowed by the presidential candidate), and 99% of adults and 99.5% of children in the Commonwealth of Massachusetts have health insurance. The initial impact of universal coverage on a provider radiology department such as ours was not that significant.

We had always had a core value of being a payor-agnostic department and would cover patients with or without health insurance. We saw a little bit of an uptick in demand, but not a tremendous difference.

Predictably, the costs borne by the state government went up—because one of the ways that we got the uninsured insured was to take everyone who was eligible for Medicaid and make sure that they were actually in Medicaid. Those who were earning up to 300% of the poverty level got state-government subsidies for their health-insurance premiums.

This is very similar to what’s going to happen on the federal level. The state government began seeing utilization creep up a little bit in all fields, including radiology also, and it saw increasing expenditures to make sure that everyone had insurance.

Now, we shift to phase two, which we sometimes call payment reform: how our insurers and our providers are responding to a climate where reducing costs and improving value are now imperative. We’ve got the coverage part; now, we have to work on the cost side.

Whelan-Downey: Coming from the for-profit market (with Shields Health Care Group), we have not seen huge upticks, in terms of volume. We have chosen to focus on a very simple concept: How much is it? A lot of our focus has been on reaching out to consumers and really helping them understand what it is that they’re going to pay for, how they’re going to pay for it, and for which part they pay (versus for which part the insurer pays).

That is actually contributing to increased volume because many other plans, as you know, are high-deductible plans. When patients are reaching into their pockets to pay for services, they want to know how much those services will cost.

We’ve been doing a lot of direct-to-consumer outreach programs through our website (and through other mechanisms), but we’re starting to see, from the IT perspective, multiple online tools becoming available to consumers. These tools try to enable them to participate in the conversation about cost (but contribute to cost containment) by understanding that if I make this choice—to go see this provider, versus that one—our total health-care cost, with my employment group, will be affected.

We still have an employment-based health-care system; employees have some control over how they’re contributing to the cost of the overall health-care marketplace, but they are also considering their individual out-of-pocket costs. Reaching out specifically to the consumer has really been our focus.

Proval: As a pioneer in the development and use of clinical decision support for radiology orders, is your department eliminating unnecessary imaging?

Seltzer: Clearly this is one element (though not the only element) of reducing total medical spending on imaging services. The decision-support system that we use was largely developed by one of our own faculty members (Ramin Khorasani, MD, MPH), who was granted a US patent on this more than two years ago. We’ve licensed the intellectual properties to a company in which our institution is an equity owner; I want to be sure that you know about that economic conflict before I comment.

The system is designed to work with an electronic health record and computerized provider order entry (CPOE) to provide advice to referring physicians at the time that they order radiology studies. The advice can be as simple as, “Great idea, let’s do it.” It can be as simple as, “Did you know that this patient had that exam last week? Are you sure you want to do it again?” It can be as simple as, “The indication was left-arm pain, and you ordered a right-arm radiograph; is something wrong with that?”

It also can be as sophisticated as, “Considering that in the last 1,000 women (aged 35 to 45) who presented with headache and no focal neurologic findings, the yield of MRI in the emergency department was next to nothing, think twice before you do this exam.” We allow physicians to override the advice that’s given, but we find that generally, they do follow the advice and the decision. 

Have we eliminated unnecessary imaging? I’m sure that the answer is no—because that is a very tall order. I wish I could tell you this is a panacea for eliminating unnecessary utilization, but I’m sure I can’t do that. In targeted situations, though, it does seem to help us look at reducing the intensity of utilization and increasing the positive yield of studies.

Clinical decision support through software is necessary—but not sufficient, we found, in managing radiology utilization. That has to be multidimensional.

In addition to clinical decision support, which we apply hundreds of thousands of times a year, we also have a peer-to-peer consultation program to get radiologists more active, more out of the shadows, and more involved in clinical decision making. Sometimes, when an order is entered into the CPOE system, the input is so unusual that we say, “We can’t even give you decision support; you’d better call Dr Khorasani or Dr Seltzer, talk to one of them, and figure out what it is that you really want to do.”

We also construct profiles of physicians’ test ordering, as a lot of insurers do. We provide educational opportunities to help them order the proper tests, and we team up with our physician organization to give them financial incentives (under the pay-for-performance systems that our insurance colleagues have developed—where typically, one of the measures is intensity of radiology utilization).

In our own organization, we’ve not put radiologists at financial risk for this because we feel that radiologists don’t order the tests. In fact, we feel as though we’re part of the solution, not part of the problem. We do try to help referring physicians earn their withholds back, however, so it’s a multidimensional program. The computer basis is amazing—and it is necessary, but not sufficient. You need other solutions as well, in our experience.

Proval: Has your imaging spending increased or decreased since health-care reform came into effect? What impact will the mandate to control cost have on your strategy?

Sherman: Imaging spending was going up to the high single-digit and low double-digit range—and again, it’s not about the radiologists. Patients don’t get up and decide to go get imaging studies; radiology is subject to demand created by others. The RBM helped to bring the trend down to about 3% a year, which is consistent with Chapter 224; our objective in this area was to ensure that referring/ordering physicians are using accepted evidence-based clinical guidelines in deciding whether to order advanced imaging studies.

We actually have exempted Partners HealthCare (which includes Brigham and Women’s Hospital and Massachusetts General Hospital) from our utilization-management program, but it’s not really about the radiologists. They know what to do. There are some other tactics that we’re using to try to keep our focus on high-value radiology procedures.

On the member side, we do have a transparency tool coming out that will provide information about the cost in different imaging facilities; we’re intending to present some quality information as well. I would suggest that if you want to be successful, either be a low-cost provider of service—or think about imaging studies not as isolated services, but as part of the larger ecosystem and as part of how you contribute to value.

Proval: There is a greater price sensitivity in today’s broader market; what role is IT playing in response?

Whelan-Downey: I find it simply astounding that patients can’t find out how much services cost. The students who are now covered, along with postgraduate students and recently graduated college students, cannot believe the lack of available information.

I employ a small group of programmers. We have a lot of think-tank discussions about how we should handle this and how we could do better. One of the things we’ve talked about doing (but haven’t done yet) is creating something like a cost-savings locator.

Maybe, if there were an app (or a site) where you could see all of the area’s providers of the relevant services, in order by distance from the patient’s home (as well as by any other factors that interest the patient), people would find that useful. I see the number of online tools increasing. I see the IT departments collaborating, between the insurers and the providers—more than ever before.

Proval: What’s your strategy for maintaining market share, with aggressive new for-profit organizations coming in—and how have your communications changed with the marketplace?

Seltzer: This is a point of distinction between the ways that an academic medical center and a high-quality community provider approach trying to retain market share. Our attitude is more focused on the value part of the proposition. We like to talk with our referring physicians, but not so much with our patients.

If you work within our integrated delivery system and use our IT tools, you’ll get advice on test selection. You may be immune from preauthorization. You can schedule the exam online at any of our facilities. You get a pretty good guarantee that the patient is going to be scanned with the most advanced imaging equipment possible.

We track radiation exposure for patients throughout our entire health-care system. We promise subspecialty interpretations for all exams. We have automated abnormal-results–reporting mechanisms that make it quick to send back actionable clinical information to referring physicians. We have an any-imaging, anywhere, anytime approach to our PACS, so any referrer having a patient’s imaging study done at any Partners provider can have that image on his or her electronic desktop with a couple of clicks.

All this happens within the electronic medical record of the Partners system, and in addition, we are now beginning to do consultation to support the care continuum. Price-based competition is bad for us, and we’re biased against it, for obvious reasons. We feel as though that also creates commoditization of our services.

We prefer to market on a value basis, so we don’t typically market directly to our patients. Our opinion is that it’s hard to understand completely the ambiguity and the obfuscation of what it really costs to get an exam. It is hard for a typical patient to make a value judgment, when there are so many factors in the value equation that he or she can’t possibly understand. For these reasons, our marketing communications are aimed directly at the physicians—in our network and outside it.

Proval: How are you demonstrating the value of your services to hospitals and payors?

Seltzer: The definition of adding value in radiology is still complicated. It’s very challenging to put a quantitative estimate on the value that a high-quality diagnostic-imaging exam provides to the patient. Once you’ve made a precise diagnosis that dictates what the treatment is going to be, savings can come from the reduced cost of therapy for a well-characterized disease.

We’re finding it harder to attribute costs and outcomes to radiology specifically, and we’re moving toward the model of making the right diagnosis and getting it exactly right—down to the genomic level. It’s interesting, it’s brand new for us, and it’s challenging.

Proval: Is Harvard Pilgrim using incentives to move radiology providers toward value-based care?

Sherman: We’re trying to put together various bundles that pay a global amount for a spectrum of care. In these emerging models, we are paying a provider group for outcomes, not volume of services—which, in some cases, could actually mean that if something goes wrong, the group might not get paid additionally for the care required to address the complication.

In these models, the physicians and other providers should not be thinking about encounters, and they are free to get away from focusing on what activities will get them paid. In order to be successful in these models, they need to be thinking about what services and tests best deliver the desired outcomes, including which services create no value (such as repeating tests that they were unaware had been done previously).

The question is not how much radiology is costing, but what the value of that episode is—and what radiology contributes. If you have radiologists who are helping to guide referrers to the right studies (or helping to prevent the wrong studies from being done) and ensuring that the information is used appropriately, that’s what I get excited about; that adds value.

Proval: How has heightened cost consciousness affected the Shields value proposition?

Whelan-Downey: Our value proposition has changed very little since the company was founded, in 1971: It’s to treat every patient as you would want your mother to be treated; be easy to do business with; and hire the best and most talented technologists and radiologists—and do it all at an affordable cost. If we do those things well, that’s how we are going to define survival ability.

We believe that there’s a place for a provider—whether it’s a for-profit organization or not—that has those kinds of values. That represents the goal: to get to a sustainable health system in which we all contribute to lowering the pooled cost.

Proval: How will Shields engage in care coordination and collaboration with other providers—and what role will IT play?

Whelan-Downey: An increasingly strategic one: All of IT and radiology, historically, has been somewhat inward. The business itself needed to run. It was an operational IT role—but with health reform, market changes, and changes in demand, we’re starting to see IT departments become more strategically imperative. You need to connect in new ways with referring physicians, patients, and even other providers in hospitals and health systems.

The collaboration component of it has probably changed form about three times in the past 30 years. In the 1990s, it was about getting everything online. In the 2000s, it became about interfaces, point-to-point connections, and passing information to one another. In the 2010s, it’s best described as (more or less) information sharing. It’s more of a platform-based approach that allows you to exchange information bidirectionally with other organizations.

This has changed the role of IT in radiology. It’s brought CIOs to boardrooms where they weren’t seen before; many joint-venture boards are starting to see vice presidents of IT included, at these strategic levels, because sharing of information is only going to grow, continue, and become more and more important. Organizations are continuing to make investments in business-intelligence solutions that are playing large roles in how they are moving forward with their various strategies.

We interface with absolutely every major hospital in the health system, in the markets in which we operate. We are passing our radiology reports/images into hospital information systems and PACS, and I think you’ll see that level of exchange increase, over time. It’s key to our survival: If we’re not providing results, in a timely way, to both the provider and the patient, that certainly wouldn’t be added value.

Proval: As more providers are pressed to assume risk, what are the implications for radiology?

Seltzer: As we move from fee-for-service payment to alternative payment methods based on risk, the implications are pretty profound for radiology. Many of the payors (as well as some providers) believe that as long as we’re in a fee-for-service environment, we will never be able to control health-care costs because we have a moral hazard: The more you do, the more you get paid.

When the Pioneer accountable-care organization (ACO) opportunity came about, five health-care systems in Eastern Massachusetts signed up to be Pioneer ACOs for Medicare. Partners is in risk contracts with all of the major private insurers in the state. Payors are moving us inexorably from fee-for-service payment to full-risk capitation. We’re not there yet; we’re halfway between those two poles, and many of our products now involve pay for performance, bundled payments, or payment against a total medical-spending budget—which is a very uncomfortable place for the radiology department to be.

We are in this very awkward, in-between situation. Let’s assume we go way over toward the risk side. What is the implication for radiology? An increasing proportion of our business is covered by risk contracts, so what would we do about it?
One answer is to continue what we’re doing, but with even more intensity: managing radiology utilization in collaboration with other providers. With our decision-support system and our other medical-management systems, we will add value to the health-care system—if we can aggressively manage radiology utilization.

A second strategy would be to keep care within the network. If patients leak out of the network, and we are fully at risk, all those technical and professional revenues that go to other providers do us no good.

How do we coordinate care with other providers? How do we team up radiologists with referring physicians in clinical service lines to decide on the appropriate algorithms for working up different clinical conditions? We have done that already. We have to do even more of that because we have to share risk with referring physicians.

We have to come out of the shadows and be consultants. The radiologist who has no exposure to referring physicians or patients is not going to fare well in a full-risk environment because he or she will be deemed irrelevant to patient care. We have to build consultative services, either in multidisciplinary clinics or by moving our reading rooms out into the clinical areas, as we are actually doing now, on an electronic leash that’s very loose.

We want to offer, in partnership with other diagnostic disciplines, the full suite of diagnostic services. If we’re getting a fixed payment, we’re going to get the diagnosis right (all the way down to the genomic level) and help the health system to save money.

Image-guided intervention is a huge opportunity for us. Here, we’re the low-cost providers, compared with surgeons—and we can provide outcomes that are just as good, with no hospital stay, less time off work, and a lower risk to the patient.

As image-guided interventionalists, we can do everything from image-guided surgery to image-guided focused ultrasound treatments for uterine fibroids. Particularly in an environment where we’re sharing a risk with our obstetrician/gynecologist friends, our incentives are, all of a sudden, aligned to get those fibroid patients treated at the lowest possible cost—because we’re in this together.

Don’t forget to push image-guided interventions as part of your solution to the problem of taking on more risk. Don’t forget that it’s just good for patient care, too. As you’re steered away from fee-for-service medicine and toward assuming more risk, remember that our ability to provide image-guided interventional services is another key capability for us. That’s what we’re doing, as we’re learning to love payment reform.


Kris Kyes is technical editor of Radiology Business Journal.