New Payment Models Driving Imaging Decision Support: Desert Radiologists

As accountable care organizations and other forms of integrated delivery networks take root nationwide, independent radiology practices are experiencing a phenomenon not regularly seen since the 1990s: capitated contracts. “The biggest change our group has seen recently is the rise in capitated contracts, where we receive a fixed reimbursement for providing imaging for a whole patient population,” says Whitney Edmister, MD, PhD, of Desert Radiologists, a 54-physician practice based in Las Vegas, Nevada. “We receive a monthly payment to provide all necessary imaging services, allowing health plans to better control their costs.”

Edmister explains that as new payment models have gained footing in the practice’s local market, health systems have been buying primary care practices in record numbers, pooling patients into their HMO networks. “If the HMOs don’t offer their own imaging services, they’ll come to us and ask whether we can provide it—along with electronic ordering and receipt of reports back into their EMR.” As a result, Desert Radiologists currently has 20 unique sites interfacing with its McKesson PACS, and is continuously adding more.

Intensive Informatics

Mike Reardon, information services director for the practice, describes the process of linking disparate EMRs to Desert Radiologists’ PACS as “intensive.” “Typically, we have to create both an incoming interface for electronic orders and an outgoing interface to provide reports back to their systems, matching up patient identifiers and medical record numbers,” he says. “One of the biggest challenges is finding a way to keep the incoming orders consistent so that our radiologists can keep reading productively.”

A worklist engine included in the McKesson PACS helps simplify the process, a boon to both radiologist productivity and a busy IT staff that needs to focus its energies elsewhere. “The McKesson PACS has really helped us be efficient,” says Teresa Grate, information systems manager for the practice. “All of these different facilities send us their information, and we can create filters inside of our PACS to customize worklists for our radiologists so that they can meet our hospitals’ contractual needs while reading from multiple sites.” Edmister adds, “Everything comes into one common worklist, so when you open up a chest x-ray and click dictate, you’re working in the same PACS environment with the same templates, irrespective of which of the 20 locations the study and prior studies came from.”

In the future, Reardon hopes the process will be simplified by a burgeoning health information exchange (HIE). “I only see the informatics side getting more complex with time as we have to interface with more and more disparate systems,” he says. “That’s why we’re working with our local HIE to start going down a different path. The HIE would allow any physician with an EMR to download our reports back to his or her own system. We wouldn’t have to built individual interfaces anymore—instead, we could upload everything to the HIE.”

Next-level Capabilities

The team at Desert Radiologists sees these changes as just the beginning—growing pains on the way to more effective management of the health of these patient populations, supported by next-generation IT. Edmister explains that the capitated contracts don’t disincentivize referring physicians from over-ordering imaging, meaning that, ultimately, they incentivize radiology practices to double down on decision support. “Because the referring physicians can order as many imaging studies as they want with no penalty to them, the risk is on us,” he says. “In order to understand the cost to us of providing these services, we have to have a very accurate accounting of what the ordering patterns are for these physicians.”

But that data can, and should, be used for much more than contract negotiation. In fact, the health systems are already mining the ordering data, and with radiology reports being pushed back into their EMRs by Desert Radiologists, they are positioned to evaluate the usefulness of an array of imaging studies. “They do a lot of data mining and population management by looking at that data,” Edmister notes. The radiology practice is already seeing changes to physician ordering patterns that reflect more emphasis on preventive care, he says: “We see greater numbers of plain films for early evaluation of things like low back pain, chest pain, and cough, and we’re seeing more patients getting plain films than CTs and MRIs.”

Ultimately, Edmister believes that the proliferation of interfaces between different organizations’ IT platforms, as well as the validation of standards across vendors, will allow further attention to appropriateness, honing the usefulness of imaging as a tool to help manage population health. “When you implement decision support at the point of imaging order entry, you have the most potential to influence the ordering of exams,” he says. “You have the opportunity to steer physicians away from low-value studies and toward studies that have the most value in terms of making the diagnosis.” Cat Vasko is editor of HealthIT Executive Forum.

Cat Vasko is editor of HealthIT Executive Forum.