A Look at Future Radiology Payment Models

What will future payment models look like in radiology? The picture has yet to come into focus, but there is one certainty: The industry, through the mandate from the Affordable Care Act, will largely move away from the fee-for-service model it has used for decades.

Industry leaders are already discussing many options, including bundled payments or payments through the accountable care organization (ACO). To offer some perspective and context on potential payment models for both inpatient and outpatient services, the American College of Radiology’s Harvey L. Neiman Health Policy Institute recently published a policy brief. Diagnostic Imaging spoke with Richard Duszak, MD, FACR, the institute’s chief executive officer, about what these possible payment models could mean and what the future might hold.

The Neiman Institute chose inpatient hospitalizations and episodic imaging with screening mammography to analyze payment models? Why?

That really gets to the core of who we are and why the Neiman Institute was founded. We’re in an environment in which payment systems are changing along with the entire healthcare delivery system. This policy brief really gets to the core of what’s probably one of the biggest pieces of the mission and our ongoing work: to try to be a thought leader for credible ways to move radiology forward in alignment with the push away from fee-for-service.

I think there’s been a huge push to move physician payments away from fee-for-service, and that’s received a lot of attention. A lot of these efforts are basically saying fee-for-service has to go away. I’m personally skeptical that fee-for-service will go away entirely, but I think its role will be dramatically diminished going forward. As we move into value-based approaches for healthcare delivery, paying physicians with an episodic - and ideally - risk-based methodology makes a lot of sense. It’s supported by a lot of people. We wanted to start laying some groundwork in common areas in which radiology services are provided. The high dollars associated with utilization on the inpatient side of things is the primary focus. The inpatient episodes are getting a lot of attention with length-of-stay with people coming up with ideas for various clinical specialties to use. Those who provide consultation and interpretative services have a less clear role under these models, and we wanted to move the conversation forward.

The second piece with screening mammography was to develop a prototype model in which radiologists could fit into a bundled payment system. This is very much in its infancy stages. This will be a challenge for a lot of imaging services - the clinical variable either happens downstream or upstream from the services provided.

But with screening mammography, it’s a situation where most of the follow-up is determined by the radiologist’s recommendation. It’s a clearer model that gets us into a service that impacts all women - roughly half of the Medicare beneficiary population. Basically, we chose these two models as a way to start the conversation and move things forward.

What information do you anticipate will be found from these inpatient and episodic analyses, and how will it inform clinical activities and [[{"type":"media","view_mode":"media_crop","fid":"12112","attributes":{"alt":"","class":"media-image media-image-right","id":"media_crop_5947385561824","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"308","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"float: right; margin: 5px; height: 176px; width: 200px;","title":"Richard Duszak, MD, FACR","typeof":"foaf:Image"}}]]decisions?

We need to start establishing an understanding of how services are provided under our current system in order to help us identify opportunities and potential problems in moving within the bundled payment environment. This is our first step. It’s an ongoing project that will keep us busy for decades. It isn’t that we’ve solved bundled payments, and we’re done. This is the ground work, and we have to look at it as a building-block approach.

With screening mammography, there’s a lot of research that has been done into the clinical side of why and when the studies are done, as well as the follow-up testing. Very little attention has been given to the financial modeling associated with that. In order to determine what the best and most cost-effective practice for negotiating a risk-bearing, reward-sharing arrangement, you first have to understand the spectrum of services provided. You have to identify the variability of where a practice fits both in the spectrum of services and geography. There’s an analogy I like to use: If I ran a marathon in X-many minutes, I don’t know if that’s good because I’m not a long-distance runner. I don’t know the variable. This is the first exercise to look at defining the soup-to-nuts accounting of services for screening mammography.

Patients coming in for a screening mammogram know there is a definite end of a presumptive benign or malignant diagnosis. A lot of times, there’s intermediate testing that frequently occurs, and when you examine what Medicare pays for their studies, you start to see where there’s a mean, median, and the top and bottom quartiles. You identify who the outliers are and why. Is the variability justifiable? Center X might be a regional referral center for high-risk patients who are genetically predisposed to breast cancer, or it could be unjustifiable, meaning providers order the tests because fee-for-services pays well in their environment.

We hope, when we match providers and their outcomes, to see if they’re providing equally good or better care at lower cost. What are they doing that establishes better practice patterns?

What do you anticipate will be the impact on provider activities and patient care?

I think, at this point, discussing what we’re going to find would be pure speculation. What we don’t know is if this will be a classic bell curve with even distribution or a fat bell curve with huge variation or a skinny bell curve. Until we are able to actually do the analysis, we will focus on screening mammography and the 700-plus DRGs for inpatient services. Each of these presents an opportunity for innovation.

I think a lot of what we’re going to see is some services where there’s pretty uniform delivery of care across the country. We need to be clear that uniform delivery doesn’t necessarily mean that it is the best delivery. There may be better technology or ways of delivering care, but if that’s how pretty much everyone is doing it with little variability, it’s probably the best we can do in 2013. Those areas would be ones that don’t get as much attention.

If we find screening mammography in a particular subgroup of low-risk patients, there could be a lot of variability in care. For example, if patients in State A are getting MRI of the breast at a 20-times higher rate than in State B, then after we do all risk adjustments for the population, we know there are two very different ways to deliver care. Getting back to the first steps in building blocks won’t tell us which is better, but it will tell us where we’re going to want to focus. We need to look at the benchmarks, the distribution, and the variability and cost of services.

What do you see happening in the future?

I see a few next-step opportunities, and this is why our work in payment models based in radiology is a big priority for us. We see this as potentially very important. First, it provides a foundation for ongoing value-based research. Once we can identify these types of variations at the national level, we can create a platform for researchers where they have patient-specific information. We’re currently using redacted Medicare claims, so we don’t know who had what, but it can provide a platform for other registries and databases, helping them move their needle forward with regard to clinical outcomes and cost-effectiveness research.

A second offshoot is that we are in an environment where increasingly at the local level, there’s a lot of discussion on the innovation of demonstration types of projects through fixed-dollar models. This is where accountable care organizations get most of the attention. What we’d like to do is have a framework in place where conversations can start to happen in communities across the country and for radiologists to be seen as real participants in these discussions.

The third area is to take this beyond the individual institution level and provide a platform at a national organization where Medicare starts talking about how to move into the bundled payment environment. Are there services that tend to go to bundled payment? What we’re trying to do is validate something that most likely in the radiology space lends itself to case-based or episode-based systems, such as screening mammography, and figure out where it works and doesn’t work.

Potentially, we could set up a platform for further investigation with the Centers for Medicare and Medicaid Services funding it with one of their innovation grants. The same holds true for any of those 700-plus inpatient DRGs that provide similar model payment where physicians receive one price for services during a patient stay. We may be able to make a good first step into moving physician payment into that arena.

The bigger picture context is everyone wants to move from fee-for-service to fee-for value. The devil - and there are a lot of them - is in the details. But we’ll start with small steps and incrementally increase risk as our knowledge base collectively moves forward.

From the Neiman Institute’s perspective, what is the take-away message?

Our key take-away message is that society and policymakers have spoken loud and clear that we collectively - all stakeholders in healthcare - need to pursue value-based payment systems rather than volume-based payment systems. Our take-home message with this brief, even though we don’t yet have the data to report now, is to say that we are in this space.

We are, to my knowledge, the only national radiology organization in this space. We are moving forward. This isn’t just all theory. We’re moving into the proof-of-concept phase, and hopefully, we’ll continue with this inertia to tangible, practical implementation payment models that at least, in certain circumstances, would be good starting places for moving radiology out of the fee-for-service into fee-for-value models should this be where the policymaking community wants us to go.