Kimberly E. Applegate, MD

A radiology department resembles an orchestra in its structure. In particular, each member must perform at a high level of expertise, and the functions of each member are highly specialized. Members cannot typically substitute for others outside their own narrow areas of subspecialization, yet each position must be filled if the group is to function properly. Any vacancy will be immediately obvious and will be detrimental to the performance of the entire group. Many radiology practices and departments now find themselves in this unenviable position as job openings for radiologists remain unfilled.

In early 2000, there were 25,600 professionally active radiologists in the United States. Each year, 1,000 to 1,100 residents finish their training in radiology and approximately 530 radiologists retire or die. The 2% annual increase in total radiologists is exceeded by the 3.5% to 5.0% annual increase in demand for radiology procedures and the annual increase in radiology relative value units being expended.1 Given the rapid expansion now being seen in consumers’ use of electronic health resources available on the Internet (see Figure 1), demand fueled by more widespread knowledge of imaging may increase even beyond the levels currently predicted. As consumers become more informed about both noninvasive diagnostic technologies and their own right to make health care decisions, they are more likely to request these services.


Carol M. Rumack, MD

Between 1995 and 2000, United States and European combined volumes increased 19% for MRI examinations, 10% for catheterization-laboratory procedures, 10% for CT studies, 8% for ultrasound examinations, and 4% for nuclear medicine procedures. Radiography volumes did not change significantly during the same period, according to the American College of Radiology.1 (See Figure 2)

As an example of the increases seen in CT workloads, one to two examinations per hour, per room, were typically being conducted in 1992, and there were approximately 40 slices acquired per study. A decade later, there were three to four examinations per hour, per room, with slices acquired per study ranging from 120 to as many as 1,000 for neck and spine CT.

In 2001, the Health Care Financing Administration projected trends that it had tracked since 1980 to arrive at predictions for 2008.2 By then, health care costs are expected to account for nearly 15% of the gross domestic product, compared with 9% in 1980. The distribution of those health care dollars will have shifted as well. The share of resources spent on long-term care will decrease slightly, from 7.1% to 7%. Costs for physicians’ services will increase from 18.3% to 19.4% of health care spending, but spending for hospital services will decrease from 41.5% to 31.4%. Spending for administration, construction, and other costs will rise from 27% to 33%. Drug costs will show the most rapid growth, from 4.9% to 9.9% of spending. These figures imply that radiology departments will still be expected to do more with less, but will they be able to function with fewer radiologists?

Figure 1. Growth in Interent use to obtain health information as predicted in 2002. Source: Deloitte Touche Tohmatsu.

Planning for radiologist recruitment and retention would appear to be needed now, before the radiologist and radiologic technologist shortages become more severe. Estimates of vacancy rates include positions that are budgeted, but not filled. According to the American Hospital Association,3 US hospitals in 2001 had vacancy rates of 21% for pharmacists, 18% for radiologic technologists, 18% for billing personnel and coders, 12% for laboratory technicians, 11% for registered nurses, and 9% for housekeeping and maintenance personnel. Is the radiologist job market changing to a similar degree? Radiologist vacancy listings in The American Journal of Roentgenology and in Radiology increased 284% between 1995 and the period from 1999 through 2001,1 peaking at 476 jobs available in June 2001. Of the openings listed during the 3-year study period, 36% were academic and 64% were in private practice. Demand was highest in pediatric radiology, mammography, and abdominal and chest imaging.

At present in the United States, 87% of radiologists are 35 to 65 years old; 5% are younger than 35 and 8% are 65 or older.2 Most (73%) work full time, but 17% are retired or not working and 10% hold part-time positions.

Fewer women would appear to be entering radiology than other fields of medicine. More than 40% of medical-school graduates are women, but only 22% of radiology residents and fellows (and 16% of all post-training radiologists) are female.4 The reasons for this discrepancy are being studied by the ACR.


According to the American Hospital Association,3 United States hospitals are taking several measures to improve their ability to recruit and retain staff. In 2001, 85% of hospitals provided tuition reimbursement, 63% offered flexible hours, 27% used bonus programs, 23% had clinical ladders in place, 22% operated under shared governance, 17% offered child care, and 5% provided transportation.

Figure 2. Diagnostic imaging procedures in Europe and USA. Source: American College of Radiology.

Radiology faculty members are typically responsive to several variables where recruitment and retention are concerned. Obviously, they want competitive salaries and state-of-the-art equipment. They also benefit from solid departmental and institutional infrastructure to support their daily work and a pleasant, safe environment in which to perform it. Their loyalty to the department can be gained by giving them reasons to belong to the group, promoting their leadership talents, and creating a sense of ownership of the department and everything that it produces. Flexibility in meeting an individual radiologist’s scheduling, task, and environmental preferences also improves the probability of retention, as does good communication within and beyond the group. In fact, a good working environment is defined by respect for each member of the staff and by effective communication with each individual. Faculty members should share, or at least understand, the department’s vision and goals. They should be made to feel respected and appreciated for their contributions to the furtherance of those goals.

The opportunity to participate in research is essential to the continuous advancement of radiology and cannot be neglected when retention and recruitment strategies are being chosen. Research is of critical importance to all radiologists, not just to those working in academic settings, because it advances daily clinical practice. Biomedical research can be thought of as a matrix, with the radiology component involving imaging, engineering, and informatics. The biology-driven component of radiology research is problem oriented, while the technology-driven component is solution oriented. Imaging science as a whole is multidisciplinary and complex, involving the work of physicians, cellular/molecular biologists, chemists, physicists, mathematicians, and engineers, and radiologists deserve the chance to be part of this research community.

The radiology department chair must understand the value of clinical activity and of research and must learn to remove obstacles that prevent radiologists from taking full advantage of research and clinical opportunities. It is helpful to design a programmatic, strategic approach that incorporates a clear definition of what the department wants to become. The department chair must also strive to pursue multiple duties in proper proportion, and to know the boundaries that apply. For example, excessive control of radiologists can be seen as an insult to their professional status, but some regulation will be needed if the department members are to work toward common goals.

The chair of a radiology department can be of great importance in the recruitment and retention of faculty. The chair must first create a shared vision and articulate goals. He or she can then set priorities for the department and deliver on the group’s commitments. The needs of the faculty must be understood, and staff training and retraining should be ongoing. A unique role or niche for the department can be created through program development, with the understanding that the department’s fiscal health and development must be tended carefully if the organization is to function at all. The work environment must be managed attentively, with an emphasis on conquering the paradox of progress: it is necessary to encourage the taking of risks while simultaneously protecting the value of continuity. Above all, the department chair must learn to love change.


If there are no changes in graduate medical education, the number of radiologists will be increasing only 1.2% per year by 2020. A reduction in residency length from 4 years to 3 years would increase the radiologist supply 1.7% per year by 2006, 2.2% by 2010, and 1.9% by 2020. Elimination of fellowships would first increase the supply of practicing radiologists 1.9% per year by 2006; however, because a lack of fellows would almost certainly become a subsequent lack of faculty to train future radiologists, the increase by 2020 would be only 1.2%.5

Through the work of organized radiology, residency slots in the United States were increased by 100 in 2001 and by 150 in 2002; the US military also added approximately 30 residency slots for each of those years. The number of finishing radiology residents is expected to increase by 63 per year. Medicare caps limit the reimbursement of teaching hospitals for the cost of training residents, but residency positions are still economically feasible because they cost less to fill than positions for other types of health care providers (especially faculty members).

The rules governing residency training in the United States affect the ability of institutions to increase the number of residency slots available. The faculty-to-resident ratio must be at least 1:1, and 7,000 cases must be available per resident. Increasing procedural volumes will therefore permit the training of more residents, if enough faculty members can be found to maintain the 1:1 ratio. Unfortunately, the current trend of faculty members leaving academic settings for private practice will decrease the ability to train residents and worsen the radiologist shortage. All radiologists ultimately will be impacted negatively by academic radiologist attrition.

Current radiology residency program requirements call for nine subspecialty faculty members. At least one faculty member must have first responsibility for each of the nine subspecialty areas, and no faculty member can be responsible for more than one area. Faculty members should spend 50% of their time in the department. Commitment to radiology training must be demonstrated by fellowship training or subspecialty practice, subspecialty society membership, publications and presentations, and annual CME in the subspecialty.


Radiology faces multiple challenges. How can the field’s growth be sustained? How can faculty best be recruited, trained, and retained? How can the diversity of roles within radiology (clinician-teacher, clinician-scientist, and researcher) be acknowledged and supported? How can uniformity in clinical care be ensured? How can standards of clinical excellence be implemented? How can a good working environment be created, especially as we promote filmless/paperless radiology?

Remote reading arrangements may be attractive, but it is important to maintain human contact with patients and referring physicians in order to prevent radiology from becoming a faceless specialty. The isolation that might result could make it difficult to keep radiologists happy with their work. Instead, faculty members should be part of a team and should seek active clinical participation wherever it is appropriate. Radiologists also can (and should) expect to become part of a hospital’s leadership over time, if they wish.

Residency applicants are increasing, but the faculty members needed to train them are decreasing. The number of finishing residents can increase only so long as faculty members are replaced by finishing fellows. Both women and men, however, are increasingly choosing private practice over academic radiology, so full replacement of retiring faculty by fellows seems unlikely.

Clearly, academic radiology departments and private practices will need to work together to craft solutions to staff shortages. Competing to lure radiologists away from academia and into private practice might ease private-practice radiologist shortages in the short term, but it only worsens the long-term problem; the more radiology faculty lists shrink, the fewer new radiologists the residency training system can produce to be available to be hired by private practice groups and the more radiology procedures may drift into the turf of other specialties for lack of staff.

Kimberly E. Applegate, MD, is associate professor and director, pediatric radiology research, Indiana University School of Medicine, Indianapolis, and Carol M. Rumack, MD, is professor of radiology and associate dean of graduate medical education, University of Colorado Health Sciences Center, Denver. This article has been excerpted from Work-force Issues in Radiology, during which Applegate and Rumack presented at the annual meeting of the Radiological Society of North America in Chicago on December 3, 2002.


  1. Bhargavan M, Sunshine JH. Workload of radiologists in the United States in 1998-1999 and trends since 1995-1996. AJR Am J Roentgenol. 2002;179:1123-8.
  2. Spring 2000 ACR survey. ACR Bulletin. 2002.
  3. American Hospital Association. Special Workforce Survey. Chicago: AHA; 2001.
  4. Saketkhoo DD, Covey AM, Sunshine J, Forman HP. Updated findings from a help wanted index of job advertisements and an analysis of the policy implications: is the job-market shortage for diagnostic radiologists stabilizing? AJR Am J Roentgenol. 2002;179:851-8.
  5. Bhargavan M, Sunshine JH, Schepps B. Too few radiologists? AJR Am J Roentgenol. 2002;178:1075-82.