The Rewards of Risk: Radiology Takes The Leap

In the billing office of a radiology practice, an employee reviews a remittance just received from a payor. Noticing that something looks different, she questions her colleague, who informs her that instead of collecting fees for individual imaging services rendered, the practice is now operating under a per-member, per-month capitated arrangement with that payor.

The leader of a hospital-based practice discusses with a payor representative the option of negotiating a withhold to be paid upon meeting utilization thresholds. The agreement absolves the practice and its referring physicians from having to jump through radiology benefits management (RBM) hoops.

Such scenarios have become reality as private radiology practices and hospital radiology departments alike gravitate slowly away from the traditional fee-for-service model and toward one in which they assume risk. Although radiology is admittedly in the early days of risk assumption, a picture of what it may look like within the specialty is beginning to emerge.

Crazy for capitation

Capitation, reportedly more prevalent in California than in other states, is beginning to take hold as a practical form of risk assumption. Los Angeles, California-based RadNet Inc., the largest provider of fixed-site imaging services in the U.S. through a network of 250 owned or operated outpatient imaging centers in several states, is a pioneer in radiology capitation. For the past 20 years, RadNet has been very active in risk assumption through capitation and delivers capitated imaging services to approximately 25 large medical groups, encompassing an estimated 800,000 covered HMO enrollees.

Capitated payments currently account for 10 percent of RadNet’s overall annual revenues and for “north of 20 percent” of annual revenues generated by facilities in California, according to Mark Stolper, chief financial officer. The majority of RadNet’s capitation contracts have been in force for more than 10 years.

Stolper says RadNet is uniquely positioned to handle risk because of its size and scale; in addition to the large number of centers, the provider has more than 400 radiologists, many of them subspecialist practitioners, whose services are exclusive to its facilities. However, successfully delivering on its risk-based contracts also has entailed the creation and administration of a comprehensive utilization management program.

At the core of the program is RadNet’s Radiology Utilization Management System (RUMS). The provider’s medical group clients utilize the system to digitally order exams and send to RadNet any additional medical indicators for more advanced imaging. Cases are reviewed for medical necessity and appropriateness according to accepted guidelines, among them nationally recognized evidence-based guidelines from multiple sources, medical literature, and consultation with subspecialty trained radiologists.

Additionally, RadNet’s utilization management department processes more than 100,000 referrals annually and leverages a staff of board-certified and state-specific licensed radiologists, physicians, registered nurses, and licensed vocational nurses. Physicians are profiled against their peers according to delivery model or specialty.

“Whether through sharing of an extensive library of articles from top medical journals or other methods, we emphasize educating referring physicians about medical necessity and appropriateness,” Stolper says. “Our focus is on offering diagnopstic tests based upon accepted criteria and guidelines.”

Meanwhile, Desert Radiologists, a large practice of 50 radiologists based in Las Vegas, maintains risk-based contracts with three payors. Capitated payments account for 25 percent of its total revenues, states Patricia A. Harms, MBA, CPA, FACMPE, the practice’s chief financial officer.

“We have not approached payors to work on a capitated basis, but we are not afraid to take on risk in this way,” asserts William P. Moore II, MBA, CRA, chief executive officer. “We anticipate that, in general, there will be a push to a risk-based model with the [advent of] ACOs and similar payment models, and our long-range goal is to be ready for that.”

Moore and Harms contend that the ease with which Desert Radiologists has transitioned into risk-based contracts through capitation has much to do with a framework that calls for payor/provider collaboration. At the outset of each contract, Desert Radiologists’ executive team meets with payor representatives once each month. After a mutually agreed-upon period of time, meetings are held bimonthly and finally, quarterly.

“We discuss clinical and business concerns, and how together, we can best serve the patients within the boundaries of cost,” Moore says. Special attention is paid to individual payors’ concerns; for instance, one participating health plan let Desert Radiologists’ management know the importance of providing “certain additional diagnostic information” to support data reported about its members. Arrangements to accommodate were quickly made.

The framework also calls for achieving specific report turnaround times, ensuring critical results reporting 100 percent of the time, and scheduling patients within specific time periods. “One of the biggest payor concerns in a capitated environment is that scope won’t expand to meet demands, or that patients” needs will not be prioritized because they do not fall into the fee-for-service classification, Moore concedes. “By the way we operate, we mitigate the concern and make it work.”

Key contract considerations

But whether a switch to a capitated model is initiated by a payor or by a radiology practice itself, groups that have made or overseen such a move emphasize that success is heavily predicated upon careful consideration of several qualifying factors to ascertain leverage (and determine whether risk-based contracting is an appropriate strategy, or if a fee-for-service model remains the more viable option.) These factors include the number of freestanding imaging sites in the market and whether hospitals that operate there are accepting lower rates to compete with freestanding sites. The percentage of outpatient imaging the practice’s sites provide in its immediate markets, as well as current patient capacity, also merits consideration.

“The greater the volume of competition from other sites and/or from hospitals that use lower rates to maintain an edge against non-hospital imaging providers, the more sense it likely makes to seriously explore the concept of risk-based contracting,” asserts Beth Scarvey, president, Healthcare Development Resources in Franklin, Tenn. Scarvey, whose firm actively manages payor contracts on the radiology side, adds that the issues surrounding outpatient imaging percentages and capacity are a bit more complex.

“The fact that radiology services have high upfront investment costs with low incremental costs makes remaining as close to capacity as possible of great benefit to practices,” she says. “However, if you gain exclusivity with a risk-based contract and your capacity is reduced, you could lose other fee-for-service patients as they take their imaging needs to centers with more capacity and flexibility in scheduling procedures.”

Other considerations also come into play here. For Stolper and his colleagues, the geographic distribution of RadNet’s own imaging centers was thought to indicate that risk-based contracting would bode well for the practice. So, too, was the breadth of imaging modalities available at RadNet facilities. “When patient lives [covered] span 50 to 100 square miles, the assumption of risk just will not work with a base of one or two imaging centers; a densely clustered group of multimodality facilities is a must in order to provide patient convenience and breadth of capabilities,” Stolper notes. Given this situation, he adds, serving large patient populations, and assuming risk in doing so, also necessitates an extensive roster of modalities, from the very routine to the very advanced.

A diligent approach to contract negotiation and management is critical as well. Several key pieces of information are needed to negotiate an appropriate rate. According to Scarvey, practices that commit themselves to serving as the exclusive imaging services provider to a payor within a particular market must obtain from that entity overall claims data for radiology utilization and membership. Such data should include CPT codes and location of services segmented by product (eg, PPO, HMO, Medicare, etc.). The per member, per month equivalent should be established and benchmarked to a fee-for-service rate, to ensure that the fee-for-service equivalent falls within an acceptable rate for the market in question.

Preparing for negotiations

In preparing for contract negotiations, Desert Radiologists reviews and brings to the table trend utilization data culled from its own systems, to see whether it coincides with what the prospective payor considers reasonable. “If we have to rely on a payor’s data to come up with the numbers, we need at least two years, especially for the more complex studies like PET and CT or MRI performed on high-tech equipment,” Moore says.

Moreover, practices will need assurance that the benefit plan design for the payor will not change for radiology services; otherwise, the volume of cases will increase significantly and possibly rise to an unwieldy level. “If the co-pay for radiology services is currently $100, and a $25 co-pay for these services is then implemented, there could be headaches ahead,” Scarvey warns. Problems also may ensue should imaging providers agree to charge-backs for out-of-network utilization, as the capitated rate collected for a given procedure will be reduced if a patient undergoes that exam at a non-network site.

Identifying services that should be carved out of the capitation comprises yet another step in the negotiation process. Procedures that involve the use of new imaging technologies or those that are performed only infrequently likely belong on this list, according to Scarvey. Moore and Harms concur, deeming radiopharmaceuticals an example of  “a sensible carve-out” Desert Radiologists has striven to execute.

“Some payors will demand an all-inclusive cap, which is a less appealing scenario since it means there is an assumption of additional risk,” Moore says. “This is something we weigh carefully. As we see it, the most important thing we—and any practice involved in capitation—can do is to establish thresholds, make certain there is every opportunity to carve out any procedures we need to be concerned about, and to put escalators in the contracts, if at all possible.” Otherwise, he says, there could be a big hit to the bottom line.

Establishing risk corridors to temper the impact on providers of capitation rates that are too low for the initial few years in which capitation is in place may prove equally worthwhile, based on the protection it would afford should utilization be higher than expected. The downside, sources say: Risk corridors do, as Scarvey puts it, limit “the upper end of profitability.”

Finally, there exist a few miscellaneous contract inclusions and assumptions to be made. For example, Desert Radiologists consents to the maintenance of utilization oversight by payors, especially when they have implemented an ACR appropriateness criteria-based clinical decision support system for directing the ordering process for referring physicians. This, Harms explains, ensures that practice and payors alike are aware that “there is some utilization management going on, utilization doesn’t run rampant, and we do not have to police it all.” Regular joint operating committee meetings are held for the purpose of closely monitoring and managing contracts, a practice Moore and Harms deem very important to the future success of any risk-based payment model.

As for assumptions, Scarvey recommends bearing in mind that the utilization of radiology services will increase for capitated providers in situations where payors are themselves transitioning from fee-for-service to capitation mode. Services that were previously performed in specialists’ offices may be diverted to provider offices where capitated rates are in effect. Additionally, she counsels practices to expect increases in imaging utilization if the payor in question had strict authorization practices in place and intends to eliminate them in line with the transition to capitation.

Beyond capitation

Still, capitation is not the answer to risk assumption for all radiology players. For Massachusetts General Hospital in Boston, sharing risk via a negotiated withhold constitutes a better approach to grappling with changes in the healthcare payments landscape.

Some nine years ago, one of the hospital’s major payors requested that it start utilizing a radiology benefit management (RBM) entity to manage utilization. This did not sit well with management or the radiology department. “We firmly believed the [procedures] ordering process was a critical component of medical management, so it should be done by our doctors, rather than by a for-profit external group,” recalls Jeffrey B. Weilburg, MD, associate medical director, Massachusetts General Physicians Organization.

At the time, the institution’s department of radiology had in place an internally developed system known as Radiology Order Entry, or ROE. An initiative to augment ROE to include an imaging clinical decision support (CDS) component—also developed in-house—was undertaken. The system, now known as ROE-DS, not only enabled the hospital to continue to meet utilization threshold requirements negotiated with its three major commercial payors—Harvard Pilgrim Health Care, Massachusetts Blue Cross Blue Shield, and Tufts Health Plan—but to keep its utilization thresholds equal to or better than utilization rates experienced by payors from other regional providers’ RBM-based solutions. In doing so, it receives a negotiated withhold, which in turn prevents the need for physicians to utilize RBM. Financial, risk-based incentives were attached to performance, allowing both the payor and the providers to benefit from optimized utilization.

Weilburg attributes Massachusetts General’s continued success in clinical performance and with sharing risk in this fashion in large part to the cooperative effort through which ROE-DS was developed. A team comprised of representatives from the department of radiology, primary care physicians, specialists, and management collaborated on the design of the decision-support rules, as well as on crafting evidence into appropriate user formats and structuring the overall design of the tool. A concerted effort was made to configure the system with clinicians’ workflow patterns in mind.

If an institution intends to use imaging CDS to support a negotiated withhold, it needs to take into account when building the tool that a surgeon is going to have a different workflow pattern than a general practitioner, and a specialist in one discipline will have a different workflow pattern than a specialist in another discipline, Weilburg says. “There has to be some customization, but not too much, yet the system needs to be powerful enough to have an impact on ordering behavior,” he adds, which does come when users have had a say in how the decision support rules look, and the formatting of the evidence—everything. It does not come when radiologists alone create the content.” Clinical and financial input from outside the department must be in the mix, Weilburg believes.

Employing ROE-DS as a teaching tool has been instrumental in keeping utilization thresholds where they should be. “Doing our own management has allowed us to gain finely grained data on the utilization patterns of our clinicians,” Weilburg shares. “We analyze this data, and provide this information to our doctors on a regular basis, so they see how their imaging utilization compares to that of their peers. This gives them information they never before had, and has sparked very useful discussion on what the optimal-use pattern should be for a given clinical situation.”

Weilburg says physicians in general like the approach. “Indeed, physician acceptance has been far higher than a requirement to get approval from the RBM of each individual imaging order,” he adds. “We believe the creation of this learning environment has been very useful in moving us toward optimized utilization and a real evidence-based medicine approach, and expect further gains are possible.”

More value-adds

Not surprisingly, hospitals and private practices alike continue to make other preparations for value-based payment as they pursue or at least contemplate risk-sharing avenues. Some preparation centers on technology refinements. At Massachusetts General, for instance, an effort is focused on improving radiology reports and the recommendations contained within them, using a proprietary tool that radiologists launch from radiology reporting stations. The tool will incorporate a wide range of algorithms and recommendations and set forth a more standardized report language. Recommendations will be wide-ranging; examples include those contained in the American College of Radiology’s series of white papers on incidental findings and Fleischner Society recommendations for management of small pulmonary nodules.

“Currently, if we were to show the same imaging exam—for instance, an exam that reveals an indeterminate mass in the liver—to 10 different radiologists, we may see multiple, very different recommendations in the report,” asserts James A. Brink, MD, Massachusetts General’s radiologist-in-chief and the Juan M. Taveras Professor of Radiology at Harvard Medical School. “Some may recommend a follow-up imaging examination while others may recommend a biopsy, some might under-recommend, while others might over-recommend. Standardization of our reports and recommendations using point-of-care implementation of diagnostic algorithms is the next big thing in adding value, and in demonstrating the value of radiology to payors.” 

Other initiatives aimed at enhancing radiology’s value proposition are linked to quality assurance and demonstrating the positive impact of radiology services on the caliber of patient care. Notably, several radiologists affiliated with Massachusetts General are now available to patients for direct, one-on-one consultations about the imaging procedures they have undergone and what course of action may subsequently be recommended for them.

For its part, Columbus Radiology in Columbus, Ohio (which favors the idea of engaging in risk-based contracts, but has yet to find an interested payor) has established a Results Communication Center arm, through which exam results are made available 24 hours a day, seven days a week. Charles McRae, Columbus Radiology’s CEO, describes the center’s “sole purpose” as “facilitating communication between referring physicians, radiologists, and technologists, and its goal, “to call every positive exam—not just critical results, but every fracture, torn ligament, and pneumonia”—in an expeditious fashion that elevates the quality of care and prevents costs from spiraling out of control.

McRae cites the example of a patient called “Susie” who had an x-ray taken on a Friday afternoon that revealed subtle right lobe pneumonia. Rather than waiting until after the weekend to contact the referring physician, center personnel do so immediately, using her preferred method of communication. “When we call the referring physician right away, that physician can in turn call Susie and tell her, ‘You have a mild case of pneumonia, and I am calling in a prescription for you that you should start now,’” McRae explains. “But if we were to wait, Susie could very well end up in the hospital with a 103-degree fever and in need of inpatient treatment with intravenous antibiotics. That is, of course, an entirely new ball game from a cost and patient care standpoint—and one payors do not like to see.”

Additionally, under the auspices of the center, Columbus Radiology holds continuing education courses for referring physicians. More than 350 physicians have participated in workshops covering such topics as image appropriateness and clinical indicators for specific imaging procedures.

Interestingly, McRae says he foresees a time when radiologists will, based on their knowledge of “which test goes with which patient” be pressed into service as imaging gatekeepers. Not all sources concur, but they do agree on one thing: Whether through capitation or another model, risk assumption, and the need to push harder to prove that radiology is of value rather than a commodity, is here to stay