SPECIAL EDITION
Cover Story

By James Brice

Imagine Dr. Rachel Young, first-year radiology resident at Hypothetical University. Over the next four years, she will hone and prove her knowledge of anatomy, physiology, and pathological processes. She will gain expertise in multiple modalities and information systems. She will emerge as a trained radiologist prepared to interpret thousands of cases annually.

Every year, academic radiology takes in hundreds of medical students like Young and turns out roughly equal numbers of qualified radiologists. The process in between centers heavily, of necessity, on imparting the clinical skills needed to meet the specialty's high standards. It also provides a crucial introduction to research and nonclinical skills-such as communication-that will prepare residents for the real world.

Gradually, internal and external pressures are modifying this well-worn path. A national move toward multiple demonstrations of physician competence calls for residents to learn professionalism and interpersonal communication skills and to begin a habit of lifelong learning. Concerns about fourth-year residents' single-minded preparation for the oral board exams has prompted a movement to postpone that particular hurdle. Meanwhile, evidence of eroding turf is prompting academic radiologists to encourage promising students to pursue research careers, beginning with exposure in residency.

What unites these and other changes is their potential to create a stronger generation of radiologists who are clinical experts, savvy in research, and attuned to the political and economic forces that shape the field. Given the daunting obstacles of surging procedure volume, a labor shortage, new technologies, and eroding turf, they may be exactly what the specialty needs.

CLASSIC FORMULA

Dr. Young's education will mirror that of residents at 196 programs around the country. Following guidelines set by the Association of Program Directors in Radiology (APDR) he will spend at least three months each in chest radiology, gastrointestinal radiology, musculoskeletal radiology, pediatric radiology, cardiovascular and interventional radiology, ultrasonography, and body CT. She may choose a four-month rotation in neuroradiology to cover neuroangiography, CT, and MRI. She will complete two-month rotations in uroradiology, mammography, body MR imaging, and emergency radiology.

She will spend countless hours at a workstation learning image interpretation and handling night call. Lectures and hands-on experience will introduce her to computer applications in radiology, practice management, health systems, and quality improvement. She will conduct research and learn how to evaluate research findings.

After her first year, she will take the American Board of Radiology's physics examination, the first of three steppingstones to board certification. In her second or third year, she will take the ABR's written clinical exam. If she follows the tradition established by countless other residents, she will sequester herself for the better part of her fourth year of residency to prepare for the oral boards.

The infamous two-day, 10-part test officially admits students into the radiological community, qualifying them for clinical practice or fellowship. The certification it imparts assures the public that the recipient is competent to practice radiology. It is also a source of untold dread.

If academic radiologists agree on any single issue, it is that something should be done to discourage fourth-year residents from focusing on the orals to the exclusion of most other activities, notably learning hands-on skills. From a faculty member's perspective, students waste their time in the library memorizing arcane facts that may appear in the test, when they could be learning more practical lessons by interacting with their instructors.

"This is the most precious time residents have," said Dr. Robert R. Hattery, president of the ABR board of trustees, referring to the fourth-year experience. "They are finally skilled enough to work hand-in-glove with faculty to learn the subtleties of what that faculty member does. You don't get that by reading a book. Yet the pressure of the impending test leads them to ignore this advice."

In April, the Society of Chairmen of Academic Radiology Departments agreed on a strategy to address this recurrent problem. It passed a resolution urging the ABR to delay the oral examination for two years after the end of a standard four-year residency program.

SCARD president Dr. Ronald L. Arenson admits that some inequities are built into the proposal. A two-year delay would keep the inevitable period of intense studying from falling into the fellowship year, but the long delay would also mean that lessons learned during residency training might be forgotten by the time a board-eligible radiologist takes the test. And community practices will probably oppose hiring radiologists who spend most of their first year studying.

Still, Dr. David S. Feigin, a professor of chest radiology at the Uniformed Services University of the Health Sciences, sees some justice in shifting the last-minute cramming to a period when board-eligible radiologists are working in private practice. Despite appeals from the RSNA and other radiology societies, group practices have been less than enthusiastic in their support of academic research, Feigin said in a question-and-answer session at the Association of University Radiologists annual meeting.

"But if we give them the responsibility for hiring radiologists who prepare for the boards while they are working for them, then each group is going to have to make some sort of systematic commitment to help their junior associates pass the boards," he said.

The resolution was submitted to the ABR for consideration by its board.

For its part, the ABR tries to keep to the middle ground as gatekeeper. It was one of the last of 24 American Board of Medical Specialties affiliates to dispense with lifetime certification. The radiology class of 2002 was the first to be awarded time-limited certificates that require additional competency testing when they expire in 10 years. Any other changes will be carefully scrutinized to assess their ultimate effects on the specialty's well-being.

"We don't want to make short-term decisions that will screw us up over the long term," Hattery said at the AUR meeting. "If we are going to make real change, whatever that is, we have to define the end process."

EMPHASIS ON COMPETENCE

During the course of her residency, Dr. Young can expect many introductions to the nonclinical aspects of medical culture. She might receive feedback about her interactions with other physicians and technologists. She might undergo role-playing exercises to improve the way she delivers clinical information to peers or patients.

Beginning this month, such activities will become far more standardized. New competency standards from the Accreditation Council for Graduate Medical Education now define what residency programs in 110 specialties, including radiology, must teach and measure. The commission began enforcing the new policies this month.

The new standards represent a shift in emphasis in that they require training and measurement of nonclinical skills, according to Dr. David C. Leach, ACGME executive director.

The APDR's education committee, chaired by Dr. Jannette Collins, an associate professor of radiology at the University of Wisconsin, Madison, hammered out a final report in January that identified training requirements and assessment tools for each skill.

A 360 degrees evaluation, for example, gives residents feedback from various observers of their performance. At the end of an interventional radiology rotation, the section chief for interventional radiology, anesthesiologists, nurses, technologists, and physician assistants will be asked to critique the resident's clinical, communications, management, and decision-making skills.

Many assessments are already a routine part of residency training, said Dr. Paul Capp, former ABR president and part of the committee that adapted the competencies for radiology. Under terms of the ACGME directives, they will be applied more formally.

"We will improve a bit by getting specific measurements from the allied health people who work in our radiology departments in terms of their professional relationships, such as how the resident relates to the rest of the team, the team being the whole department," he said.

The competency standards will be implemented in phases through 2015, Leach said. Beginning this month, programs that are due for recertification must show a good faith effort toward applying the criteria.

If Dr. Young wants to get a head start on the new requirements, she might spend some time with the APDR-ACR nonclinical skills video program. Created in 1999, it is an inexpensive way to cover issues that do not directly involve image interpretation, said Dr. Philip Alderson, chair of radiology at Columbia University and head of the committee that designed the program. Topics include job search and contracting, practical business issues, ACR standards and accreditation, critical thinking skills, ethics, interpersonal and communication skills, medical organizational politics, clinical trials, and radiation biology.

Seven lectures and a roundtable discussion were recorded before audiences of radiology residents in Bethesda, MD, and Washington, DC, between July 1999 and August 2001. A videotape of the lecture, Microsoft PowerPoint slides, pre- and post-tests, and teaching instructions were mailed to all diagnostic radiology residency program directors after each event. The series also can be accessed on the ACR Web site (www.acr.org).

Some residency programs have already begun to apply the six competencies' philosophy. The first few days' training in Indiana University's residency program emphasizes communication skills and ethics, said Ronald E. Fraley, education operations manager. Through role-playing, residents learn proper and improper responses to angry, emotional, or confused patients. Instructors also teach the essentials of proper report dictation and ethical professional behavior.

"The fact that we take time at the very beginning of the residency for these issues legitimizes them as subjects of real importance," he said.

Residents at Brown University learn how to use the ACR appropriateness criteria from lectures prepared by their classmates, said Dr. Martha B. Mainiero, resident program director. During the academic year, each resident delivers about four lectures that consider how the criteria apply to specific clinical situations. A faculty member who specializes in the topic of the day critiques the presentation and adds supplementary information about the most cost-effective approach to rendering the diagnosis.

BETTER INTERVENTIONALISTS

If Dr. Young decides that she wants to pursue interventional radiology, the path varies, and she may need to seek exemptions in some rotations to gain the training time she needs in other areas. It has been a frustrating process for budding interventionalists, but a clinical pathway approved by the ABR in 2000 may make it smoother.

The 60-month program leads to board certification in radiology and a certificate of added qualification in interventional radiology. Twenty-nine months are devoted to diagnostic radiology; 24 months, including a fellowship year, are designated for IR. Four months are spent in a nonradiologic rotation that could include experience in an intensive care unit or anesthesia training, and three months are dedicated to research.

The Society of Interventional Radiology has also been pushing the ABR to sanction a proposed 2-2-2 program, according to Dr. Anne Roberts, chief of vascular and interventional radiology at the University of California, San Diego. It would consist of two years of clinical training, two years of radiology residency, and two years of IR education. As with the new IR pathway, physicians who successfully complete the programs and associated ABR exam would receive radiology board certification and an IR certificate of added qualifications.

The 2-2-2 program would attract second-year surgical or cardiology residents who wish to switch to IR, Roberts said. She herself completed two years of ob/gyn residency before shifting to IR.

"That second year of surgery training can count toward their radiology residency. They would then spend two years learning radiology and two more years doing IR," she said.

The 2-2-2 program would also expand the pool of potential interventional radiology fellows to include medical residents who are more interested in minimally invasive surgery than most diagnostically oriented radiologists, according to Dr. Frederick S. Keller, director of the Dotter Interventional Institute. The initiative would help fill vacant fellowship slots and attract physicians with fresh perspective into interventional radiology.

CREATING A RESEARCH CULTURE

At some time during her training, Dr. Young will participate in a research project that will provide at least a basic introduction to a completely different life: that of a research-oriented radiologist. If she enjoys the work and shows promise, she may pursue other projects. With mentorship and encouragement, she might decide that an academic career is right for her.

Finding and nurturing such prospects has become a crucial mission for academic radiologists. Radiologists have been concerned for years about the relatively poor quality of their research compared with other medical specialties. Critiques of the radiological literature fault its researchers for relying too heavily on descriptive types of scientific investigations that hide biases and statistical errors.

Speaking at the 2002 AUR annual meeting, Dr. R. Nick Bryan, radiology chair at the University of Pennsylvania, saw merit in descriptive research, which he likened to Charles Darwin's journalizing in The Voyage of the Beagle. But he stressed that it lacks the methodological power of hypothesis-driven science, which is the core of Darwin's The Origin of Species. Descriptive research has actually served radiology well, Bryan said, but a shift to more disciplined methods is important.

The Human Genome Project and other advancements in molecular biology are driving the need to improve radiology research. The pace of scientific discovery has accelerated because computer-assisted biochemical assay techniques can sort through millions of molecules in hours to identify the best therapeutic candidates. Disease will ultimately be treated by altering the composition of genes or proteomic behavior of cells. According to various academic sources, radiology's scientists must become more sophisticated in their application of basic science if medical imaging is to develop modalities that can complement these new molecular therapeutic technologies.

Productive research is essential to protecting radiology turf, said Dr. Stanley Baum, former chair of radiology at the University of Pennsylvania. Research that may be a financial drain on individual departments can be a clinical boon to the field as a whole. PET, for example, gained popularity only after 30 years of scientific nurturing at a few academic centers such as the University of Pennsylvania.

"We supported PET because if we didn't support it, the dean would have given it over to neurology," Baum said.

Basic changes in the culture of radiology are required to improve the quality of imaging research, according to Dr. Robert F. Mattrey, a professor of radiology at the University of California, San Diego. Many of the pieces required for positive change are in place. Radiology now has an advocate at the National Institutes of Health, with the confirmation of Dr. Elias Zerhouni as new NIH chief. Radiology also has its own NIH institute, the National Institute of Biomedical Imaging and Bioengineering, and the federal government is spending hundreds of millions of dollars on cancer imaging research.

"We have a huge opportunity, perhaps the best opportunity in many years, as a specialty," Mattrey said.

As reflected in a plenary session at the AUR annual meeting, the academic research community is defining the terms for reform. Bryant said that resident recruitment must be modified to attract students who wish to pursue research careers. Men and women who have completed doctoral level training in basic sciences such as chemistry or cellular biology must be persuaded to add to their education with residencies and fellowships in radiology.

"People with a general interest in research won't cut it. They must be driven from the very beginning of their university training to become serious scientists," he said.

Students over 30 are too old to have the energy that this work consumes, according to Bryant. Candidates who have accumulated more than $100,000 of debt for previous schooling will probably wash out. Academic radiologists will rarely earn as much as their clinical colleagues, so materialists need not apply. Marriage and children could dull the focus, Bryant said, although Dr. Etta Pisano, a professor of radiology at Duke University, argued that radiology departments should not discriminate against potential candidates who crave conventional family living.

"Research can offer a wonderful career for people who have families. It is more flexible than clinical work by far," she said.

Radiology departments can retain their most innovative researchers with money, time, and recognition, said Dr. Carl E. Ravin, radiology chair at Duke University. Department heads must strive to achieve parity in compensation between research radiologists and their clinical counterparts.

"If your starting salary for junior fellows is $150,000, your starting salary for researchers also needs to be close to $150,000. It has to be at a comparable level," he said.

Income inequities can be balanced by providing academic radiologists with protected time for research, Ravin said. Few research careers have blossomed from science performed one or two days a week, so at least half of an academic radiologist's time should be set aside for research.

Department chairs should also find ways to recognize their researchers' successes. Social and educational mechanisms ought to be established to encourage interactions between Ph.D.-level staff and practicing radiologists, he said. Researchers should be included in grand rounds and internal conferences to showcase their work. Simple mechanisms such as bulletin board displays can celebrate the staff's contributions to professional conferences and peer-reviewed radiology literature.

BLUEPRINT FOR REFORM

A blueprint for reform of academic radiology was drafted in July 2001 during a summit meeting in San Francisco of the 40 academic radiology programs that receive the most NIH research grant support. The meeting, informally supported by the Academy of Radiology Research, proposed a new three-tier structure for academic radiology to increase the number of clinical scientists in the specialty. Dr. Bruce Hillman, radiology chair at the University of Virginia, was the main architect.

The plan was largely modeled after surgical training requirements, which mandate up to two years of bench research for surgical residents, said Dr. Haile Debas, dean of the School of Medicine at the University of California, San Francisco, who addressed the meeting.

The three-tier proposal is based on the understanding that most radiology training programs can provide all residents and fellows with significant research experience, but few possess the lab resources to carry out hypothesis-based research, according to Hillman. In practice, residency programs would be asked to declare their status.

Only about 10 medical schools in the U.S. would qualify for the first tier, Hillman said. The facilities must already rank among the top NIH-supported radiology departments. Their animal, engineering, or clinical laboratories should be large enough to give residents one-on-one faculty guidance in the intricacies of research.

All trainees would devote at least one year to developing lab, engineering, or clinical research skills. Residency and fellowship education would be geared toward securing NIH R01 grants that sponsor several years of research. Students would be taught how to qualify later for grants that will make their work financially self-sufficient, he said.

UCSF, the meeting's host, qualifies as a tier-one school, in part because of its ranking among the top five academic radiology departments for NIH funding, according to radiology chair Dr. Ronald Arenson.

The infrastructure that is possible at UCSF because of its NIH success enables the program to support 12 animal research laboratories. Many UCSF radiology residents and fellows attend epidemiology and biostatistics courses. The radiology department employs a biostatistician who advises the research staff.

"We have a whole infrastructure in place to support research. There is faculty available to mentor our residents and fellows and involve them in teams to work on projects," Arenson said.

Tier-two facilities are not as strong as tier-one schools but are capable of engaging in important research. They have enough research resources to support the training of one or two world-class researchers in each class, Hillman said. They would receive the same kind of training as their tier-one peers. Other tier-two trainees would receive at least three months of required research experience and a course in critical thinking.

The University of Virginia exemplifies a tier-two program, he said. For each of the past nine years, a single resident has been selected for individualized research training. The second year of his or her residency is devoted to research experience and training in clinical or laboratory tracks that lead to a master's degree in science. These exceptional students are assigned a research mentor, with whom they work on a major research project over that year's time.

Two of the four students who have completed their fellowships have announced their intention to continue with a research career.

Otherwise, all University of Virginia clinical residents receive one month of required research experience. Four-fifths of the students publish a paper based on that exposure, Hillman said. Residents learn critical thinking skills during lectures, class discussions, and participation in a mandatory journal club. Several articles are evaluated in each session.

Most radiology residency programs occupy the third tier. They may not have sufficient resources for hypothesis-based research, but they are well qualified to prepare residents for general and subspecialty clinical practice. Their mission includes teaching fundamental research skills and critical thinking to their residents and fellows.

"These people absolutely must come out of their training with the ability to understand when to believe in the presentation of a scientific paper and to identify biases when they read a journal or report," Hillman said.

3-2 PROGRAMS

Other universities have adopted a 3-2 configuration to give radiology residents more research exposure. After completing a three-year residency, Duke University radiology students proceed to a fellowship year and a year of resident research, Ravin said. The Duke program has attracted more research-minded residents since the change.

"There was a lot of pushback from the faculty, more because they loved the kind of residents we were getting and they didn't want to change to something else. More surprising was the pushback from the residents. They were entirely opposed to doing research. They didn't want to become geeky researchers, but we have had our best classes ever since going to a 3-2 system," Ravin said.

Many academic radiologists agree that the core residency curriculum can be learned in three years, Arenson said. They disagree about how that extra year should be used, however.

"Many of us believe the residents need additional training such as a two-year fellowship for subspecialization and research," he said.

PREPARING FOR MOLECULAR AGE

Department chairs and program directors are just beginning to come to grips with how resident training should be modified for the coming molecular age of clinical practice. Some leaders, such as Dr. Thomas Brady, associate chair of radiology research at Massachusetts General Hospital, believe it is still too early plan for those changes.

"Today's residents would forget what they've learned before they have a chance to use it," he said.

Dr. King C.P. Li, associate director for radiology and imaging science at the NIH, counters that residents should be taught now about the genetic markers of cancer because those markers are already factored into the design and selection of therapy.

At the University of California, Los Angeles, Dr. Dieter R. Enzmann, radiology chair, and Dr. Sam Gambhir, director of the Crump Institute for Molecular Imaging, are determining how to train the first generation of residents who will specialize in molecular research. Customized programs for these pioneers include experience in UCLA's molecular lab and an emphasis on nuclear imaging, Gambhir said. They will learn the fundamentals of cell and molecular biology: how to insert DNA in a cell, how to design molecular probes, and how to perform small-animal experiments using micro-PET, micro-CT, MR, and optical imaging.

This first generation of molecular imaging researchers will ultimately develop the applications that radiologists will apply to clinical practice in the future, according to Enzmann.

"We are entering a new era in terms of research, but we are not quite at the stage where we are ready to train molecular imagers. That may not happen in four years or even eight years. We must create the field before we can teach residents to use it," he said.

Taking the long view seems to be the essence of building better radiologists. At the end of her training, Dr. Young should be able to provide excellent interpretations of clinical images. Equally important for her own success and that of the specialty, she should be able to work effectively and efficiently in the modern healthcare arena, and understand and contribute to the scientific advances that will secure radiology's future.

ACGME standards for residents

MEDICAL KNOWLEDGE

Residents must demonstrate knowledge about established and evolving biomedical and clinical sciences and the application of this knowledge to patient care.

PROFESSIONALISM

Demonstrate a commitment to carrying out professional responsibilities, adherence to ethical principles, and sensitivity to a diverse patient population.

PATIENT CARE

Provide patient care that is compassionate, appropriate, and effective.

PRACTICE-BASED LEARNING AND IMPROVEMENT

Residents must be able to investigate and evaluate their patient care practices, and appraise and assimilate scientific evidence in order to improve their radiologic practices.

INTERPERSONAL AND COMMUNICATION SKILLS

Residents must demonstrate interpersonal and communication skills that result in effective information exchange with patients, patient family members, medical students, other residents, supervising faculty, referring physicians, technologists, nurses, and other members of the healthcare team.

SYSTEMS-BASED PRACTICE

Demonstrate an awareness and responsiveness to the larger context and system of healthcare and the ability to effectively call on system resources to provide optimal care.