The Future of Radiology: How to Thrive in 2012 & Beyond

A few short years ago, a radiologist's job description was relatively straightforward. Radiologists were expected to efficiently review and report on medical images. Many opted to segregate themselves in the reading room. Some outsourced night and weekend call. Few deigned to serve on hospital committees. Most enjoyed professional success. Fast forward to the brave new world of 2012 and beyond.

Performance expectations and complexity have multiplied exponentially. Radiologists are expected to utilize informatics to deliver timely reports enriched with clinical content; develop meaningful relationships with hospitals, referring physicians and patients; manage imaging utilization; participate in strategic planning and engage with quality initiatives. And more. It's a tall order.

Experts agree: In the next few years, there will be winners and losers among radiology practices. "In the past, everyone was a winner. It took work to fail. Now, it's going to take work to succeed," opines Lawrence R. Muroff, MD, CEO and president of Imaging Consultants in Tampa, Fla.

"The days when a radiologist could say 'I'm a good radiologist and that's all that matters are over.' Radiologists are expected to be good. It's a given," Muroff continues. Radiology practices are going to have to master a host of new terms, concepts and responsibilities. Service, quality and value-added top the list.

"The American Board of Radiology examination will no longer be the definitive imprimatur of quality and value. Rather, radiologists must provide additional evidence to hospitals and payors that what they do is valuable," wrote Eugene C. Lin, MD, from the department of radiology at Virginia Mason Medical Center in Seattle, in the October issue of American Journal of Roentgenology.

The hitch is determining what value-added services are necessary and how to effectively and efficiently deliver them.

Future-ready Radiology: 5 Pioneers

As most practices grapple with the how’s of future practice models, a handful of pioneers have blazed forward. Take for example: Millard Fillmore Gates Hospital in Buffalo, N.Y., has employed a CT-based stroke triage protocol that informs personalized treatment and saves $762,000 annually. Instead of applying a hard time rule about a patient’s eligibility for treatment, the neurendovascular team bases treatment decisions on CT perfusion and CT angiogram data. The proportion of stroke patients discharged home rather than to a nursing home has climbed from slightly more than 30 percent to greater than 50 percent. In the wake of the U.S. Preventive Services Task Force's revision of screening mammography guidelines in 2009, Michael N. Linver, MD, director of mammography at X-Ray Associates of New Mexico in Albuquerque, N.M., recognized the value of patient communication and clinical leadership. He crafted a patient letter outlining the ramifications of the guidelines and the role of screening mammography. Linver also provides education for referring clinicians through lectures and webinars, and shares relevant screening mammography success stories with physicians via personal phone calls. Vermont consistently ranks among the lowest in per capita imaging utilization. Fletcher Allen Health Care in Burlington, Vt., employs a two-pronged approach to curbing utilization and provides radiologists with the tools to reduce over-utilization. The EHR presents radiologists with a complete record of prior tests and imaging studies, and the radiology department has carefully cultivated relationships with referring physicians to help take the sting out of imaging suggestions. University of Chicago Hospitals devised a closed-loop imaging project that automates the CT protocol process, boosting quality and efficiency. The model cut set-up time from between five to 20 minutes to less than one. Automating the protocol process also enables community hospitals to offer complex protocols without negatively impacting efficiency. The end results are compelling. For a standard CT exam, such as an abdominal pelvic CT, the closed-loop model produced a 66 percent reduction in technologist and room time, from 30 minutes to 10 minutes. After Inland Imaging of Spokane, Wash., conducted research on local women, age 40 to 65, and discovered large pockets of women unscreened for breast cancer, the practice launched a web-based campaign targeting more than 3,000 Facebook members to fuel participation in screening. In economically sparse 2009, Inland saw a 9 percent jump in mammography and a 14 percent growth in cancers found—owing to the jump in under-screened women.

Radiology 2015

Imaging circa 2015 may not be all that different from imaging 2000, says Paul J. Chang, MD, vice chair radiology informatics and medical director of enterprise imaging at University of Chicago Medical Center. He identified key challenges confronting radiology during a keynote address delivered in 2000. At the time, he pegged increasing complexity and size of imaging datasets, the need for better integration among IT systems, increasingly complex and distributed healthcare delivery systems and evolving user expectations and economic constraints as key drivers.

However, he doesn't think much has changed. "[Today's issues] are similar to 2000, but they are even more challenging," Chang says.

The complexity in datasets now extends beyond imaging modalities, as the advent of proteomics and genomics has intersected with personalized medicine. "We have to tailor imaging to more personalized treatment options," says Chang.

Developments in molecular medicine aren't the only factor fueling this expectation. Pay for performance and meaningful use requirements also translate into increased demands for radiologists to interact with patient data and information systems. This paradigm shift means radiologists need to think in terms of managing the patient rather than managing images.

This expectation ups the demands on informatics systems and requires systems to present radiologists with patient history and clinical context. Radiologists need to go above and beyond mere image interpretation and focus on patient management.

Clinical partners

In the past, a radiologist might "get away" with interpreting a study with a mass in the liver as a "possible abscess or tumor: clinical correlation suggested." This is no longer acceptable in today's world, says Chang. Radiologists, to demonstrate their value, need to be true clinical partners and incorporate the patient's clinical context in their interpretation.

Informatics could facilitate this partnership model. For most radiologists, clinical context is trapped in the EMR. It might take five to 10 minutes to log into the system and wade through mountains of data to find key information such as laboratory values.

The PACS of the future needs to be a patient-centric system with "information" hanging protocols that concisely present key data, such as patient problem lists, lab values, prior recommendations for imaging and a timeline of surgeries and admissions, shares Eliot L. Siegel, MD, director of radiology at Baltimore Veterans Affairs Medical Center.

For example, at University of Chicago, Chang and his colleagues have developed software to provide clinical data within PACS. With access to those data, the radiologist can report the mass in the liver and add "given the elevated white blood count and fever, abscess is more likely than tumor." This allows the radiologist to demonstrate value as a clinical partner, says Chang.  

A host of other trends also are fueling the partnership model. Declining reimbursement will continue, and alternative payment mechanisms are likely to expand, says Muroff. "Fee-for-service is not going to be the sole means of compensation for radiology in the future. It may not even be a major source of compensation. Radiologists are going to have to learn to participate in alternative payment mechanisms in a manner that will be beneficial to them and fair to other physicians."

Strategic planners

The ultimate forms of alternate compensation remain hazy, but there are some common denominators among accountable care organizations (ACOs), capitation and bundled payments.

Muroff recommends two primary strategies to prepare for alternate compensation. Radiology practices need to become indispensable in their local setting. Nontraditional providers, specifically teleradiology companies, offer many advantages, including 24x7 coverage, subspecialty expertise and quality metrics. "These are legitimate needs for hospitals and patients. Radiology groups have to provide similar or better service," says Muroff.

Better service means clinical partnership, a value-added radiology report informed by clinical data embedded in PACS. Clinical partnership takes multiple forms. Radiologists should integrate themselves into the medical, social and political structures of hospitals by serving on committees and holding medical offices. "This is the only area where a local group has an advantage over a national company," asserts Muroff.

"It's very important to maintain a good relationship with [local] hospital systems," adds Fred Gaschen, MA, executive vice president of Radiology Associates of Sacramento (RAS) in California.    

The challenge for radiology, observes Chang, is to achieve the optimal balance between commoditization and value. Escalating costs make it impossible to ignore inefficiencies and wasted resources in healthcare. Commoditized radiology services may drive down costs, but ultimately are likely to deliver poorer quality, Chang says.

Smart informatics could wed efficiency, quality and value. For example, tighter integration between RIS, PACS and other systems that extracts information from the EMR boosts efficiency by eliminating the need for the radiologist to search for clinical data while enhancing patient safety, value and clinical collaboration.

Another piece of the puzzle is governance. An effective governance structure authorized to make decisions effectively, appropriately and efficiently differentiates great radiology practices from the pack, says Muroff.

Muroff recommends that practices engage in a pair of scenario planning exercises. First, the group should devote planning time to crafting a response to a hypothetical hospital request for participation in an ACO or bundled payment program.

Gaschen, whose practice survived and thrived in the capitation capital of the U.S., notes that alternative compensation is not necessarily a negative proposition. "There are advantages to capitation. It depends on how strong and smart the practice is," he explains.

According to Gaschen, a practice can leverage capitation to gain the upper hand in turf wars. Take for example a practice that has lost interventional radiology or cardiac procedures to another specialty. "Negotiate these procedures into the capitated contract," he recommends. By doing so, the practice can regain lost procedures.

Once again, informatics reigns supreme as the practice needs to pinpoint its opportunities. Such data should be stored and available in the practice management system. "Make sure you can access data quickly and slice them any way needed," recommends Gaschen.

Muroff recommends a second-scenario planning exercise and says groups should consider how they would respond to a hospital plan to post a request for proposal for radiology services. Proactively considering this question allows practices to assess and buttress their weak points.

Imaging gatekeepers

For decades, radiologists have adapted to the fee-for-service payment model by honing hyper-efficiency to a science. The formula was simple. Reading more studies meant more income. It's an attractive, but unsustainable, model.

"We are no longer going to be able to thrive by reading images. We have to manage images, including appropriate utilization of images," states Chang. This new responsibility requires radiologists to embrace computerized physician order entry and decision support.

Again, this seemingly seismic shift may be an opportunity for radiologists. "One of the potential advantages of an ACO model is that the radiology practice could articulate a value proposition and suggest a capitated model where radiologists are paid not only to read studies, but also as a broker of imaging," says Chang.

RAS polished that approach with capitated practices. The practice shared imaging guidelines with referring physicians and provided practices with quarterly reports on physician utilization. Reports focused on each individual practice and compared physicians with similar providers in the organization to help identify outliers. The program required RAS radiologists to annotate each report with a note stating whether the referral was appropriate or not.

RAS' quality control physician met with one practice twice weekly to review requests for advanced imaging studies. "Our quality control physician denied a fair number of inappropriate requests," says Gaschen, adding that the model succeeded for two reasons. RAS had developed strong relationships with referring providers, and its radiologists focused on clinical service to physicians and patients, while administrators handled financial aspects of the practice.

There's very little incentive for radiologists to curb imaging utilization under the fee-for-service payment model. However, under alternate payment structures, radiologists may be incentivized or required to assume the gatekeeper role. Wise practices may consider this scenario in their strategic planning sessions.  

Getting ahead of the game

Disaster and opportunity are flip sides of the same coin. Impending reimbursement cuts and alternate payment mechanisms coupled with increasing competition will change the practice of radiology. Success won't be a given, but it is possible, and likely, for those who prepare for it. It requires radiologists to embrace new responsibilities, commit to internal planning exercises and demand the right tools from vendors.

Siegel waxes optimistic. "Based on our last 18 years of experience with PACS and soft-copy interpretation, advanced visualization and speech recognition, we've seen radiologists figure out creative ways to maintain productivity, while providing additional value." He encourages radiologists to challenge PACS vendors with new ideas and suggestions, an approach Chang echoes.

Radiologists, Chang says, need to think about the tools that enable them to prosper and improve patient care, rather than the next iPad app. "We need [to work with vendors] to re-engineer workflow to help us stay relevant, engaged and aligned with the enterprise. It will require deep, sophisticated integration."

Delivering quality, value-added radiology services has become a business essential. Practices that adhere to hyper-efficient, volume-based models will be challenged to survive over the next decade. Those that proactively plan for and adjust to the evolving healthcare delivery system, on the other hand, are likely to find success.

Goodbye to the Text Report?

Over the next few years, radiology informatics researchers at Massachusetts General Hospital (MGH) in Boston aim to overcome the decided lack of progress in radiology reporting and launch a visual representation of the patient’s imaging history. The work-in-progress, dubbed NETRA for Navigational Enterprise Tool for Representing rAdiology reports, concisely incorporates all medically relevant data, such as a pulmonary nodule or liver cyst, and automatically maps it to a customized model of the patient, explains Supriya Gupta, MD, MGH fellow in imaging informatics.

Graphical Sketch of a Patient's Radiological History: NETRA provides an at-glance, easy-to-reference, personalized encaspulation of a patient's imaging history, including relevant prior images and data.

Gupta, who is developing the system with Keith Dreyer, DO, PhD, MGH’s vice chairman, radiology computing and information sciences, outlines a few early applications of NETRA. The system may supplement the traditional text report and provide a much more efficient way for radiologists to locate patient data. For example, under the current text report model, when a radiologist needs to determine if a liver cyst was present on prior exams, he or she has to dig through previous imaging records and retrieve and review abdominal studies to search for the liver cyst. In contrast, NETRA supplies an at-a-glance answer. NETRA also displays radiation exposure by providing a graphical representation of previous CT exams as well as an estimate of cumulative exposure. “We can assess individual risk without having to manually review the entire record,” says Gupta. The system could streamline follow up of suspicious nodules, allowing radiologists to quickly review changes in pulmonary nodule size and appearance over time. NETRA, notes Gupta, could facilitate patient-centered radiology. Without direct patient contact, it is difficult for radiologists to convey findings to patients. NETRA shows findings, such as a liver cyst, in a simple format.