| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Yale University School of Medicine, 333
Cedar St., SP2-332, New Haven, CT 06520.
2 Department of Diagnostic Radiology, Memorial Sloan-Kettering Cancer Center,
1275 New York Ave., New York, NY 10021.
3 Research Department, American College of Radiology, 1891 Preston White Dr.,
Reston, VA 20191.
4 Department of Economics, Yale College, Yale University, New Haven, CT
06520.
5 School of Management, Yale University, NewHaven, CT06520.
Received March 8, 2002;
accepted after revision March 29, 2002.
Supported by the American College of Radiology.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. All diagnostic radiology jobs advertised in the American Journal of Roentgenology and Radiology between January 2000 and December 2001 were coded by practice type, geographic location, and subspecialty and were compared with the previously published results from 1991 through 1999.
RESULTS. From January 1999 through December 2001, 15,205 positions were advertised for diagnostic radiologists, representing a 284% average per-month increase as compared with the previous 4-year period. The 12-month rolling average of job advertisements peaked in June 2001 at 476 and has since stabilized. Thirty-six percent of positions advertised were academic, identical to the proportion found from 1995 to 1998. A statistically significant relative increase in jobs advertised was noted in the Midwest, and relative decreases were seen in the Northeast, Northwest, Southwest, and (most recently) California. Statistically significant relative decreases were also observed in the number of general radiology, vascular and interventional, and emergency radiology positions. The demand for mammographers, pediatric radiologists, neuroradiologists, abdominal imagers, and chest radiologists all exhibited statistically significant relative increases.
CONCLUSION. The absolute demand for both private and academic radiologists continued to grow throughout the country and in all subspecialties, but the pace of increase has slowed dramatically during the past 12 months, especially in the western United States. Current policy should be directed toward training for the areas of greatest need.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Residency director and graduate surveys are currently the most common method used for assessing the radiology job market. The surveys are well-suited for depicting past market trends but are limited in that they present a unilateral, supply-side (job-seekers') outlook on the market and provide retrospective information but not concurrent or predictive information. Help wanted indexes, such as the one used in this study, can serve as a method of obtaining accurate data about these current and future trends [5,6,7,8] and present information from the demand-side (employers') perspective of the job market.
Accurate and detailed information about supply and demand imbalances in the radiology job market yields great value to government policy makers, radiology residency programs devising curricula, recent graduates seeking jobs, radiology groups interested in hiring, and even medical students planning careers [9]. For example, Anzilotti et al. [10] recently showed that changes in radiology hiring trends directly affect medical students' decisions to select radiology as a specialty.
We have previously shown the validity and value of these help wanted indexes [9]. This study presents the most current data on the diagnostic radiology job market from January 2000 through December 2001, obtained from our help wanted index, and discusses how this data may influence policy planning.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
"Type of practice" was divided into academic versus private categories. Any private practice or partnership-track position was listed as private, whether or not there was a university affiliation. All faculty positions, positions with academic ranking, and Veterans Administration medical center positions were considered to be academic. Administrative jobs were listed either as private or academic depending on whether they were associated with private groups or with academic centers. Nonradiology and locum tenens positions were not counted, but part-time positions lasting for over 1 year were included. As before, no attempt was made to screen out repeated ads. Neither nationwide advertisements, which offered multiple jobs across the country within the same listing, nor purely graphic advertisements were counted, all consistent with our previous methodology [9, 11].
For coding geographic region, we divided the United States into six parts: the Northeast, Midwest, Southeast, Northwest, Southwest, and California. The states included in each region are shown in Figure 1. Jobs advertised in Canada were also included in our data, but other foreign advertisements were not.
|
Subspecialties included were general radiology, mammography, abdominal and cross-sectional imaging, vascular and interventional radiology, neuroradiology, pediatric radiology, chest radiology, musculoskeletal radiology, emergency radiology, nuclear medicine, and "other." Nighthawk positions were coded as general, part-time, or other, depending on how each was specifically advertised. As in our prior articles, abdominal and cross-sectional radiology encompassed CT, sonography, MR imaging, gastrointestinal imaging, and pelvic radiology. Subcategories of abdominal imaging included MR imaging and sonography and were tabulated as a basis for future comparison in these potential growth areas. "Other" consisted mostly of administrative, chair, research, computer-interactive, and miscellaneous positions. A single position that was advertised as being split evenly between two subspecialty categories was entered as a half of each position, whereas single positions that involved three or more subspecialties were coded as one general position. From this point forward, the terms "positions" and "ads" will be used interchangeably to describe specific diagnostic radiology jobs that were coded and included in our study.
The collection, presentation, and statistical analysis of all the data were performed in a similar manner to our previous two studies and are described in detail in the original article [9, 11]. In interpreting the results, we pooled the new data from January 2000 to December 2001 with the data for 1999 from our most recent article [11]. Pooling of new and old data was performed merely for purposes of statistical comparison, allowing a larger 36-month sample period to be compared with the two previous 48-month periods (1991-1994 and 1995-1998). The actual current data, month by month, are presented in the tables and figures.
In addition, to determine the locations of inflection points in the 12-month rolling average of ads per month, we calculated and plotted the percentages of change in the number of ads published from month to month. Inflection points were determined by noting positions at which this function crossed the x-axis. When comparing geographic location and subspecialty data across time periods, we performed a statistical analysis using a two-tailed comparison of proportions. P values less than 0.05 were considered to be statistically significant; p values comparing the data from 1999 to 2001 versus those from 1995 to 1998 are reported in Tables 1 and 2.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
When the percentage change in the rolling average data from month to month is calculated (equivalent to the first derivative of the rolling average plot), three clear inflection points can be identified (Fig. 3). In June 1992, the percentage of change switches from positive to negative; in December 1995, the percentage of change switches back from negative to positive; and in March 2001, the percentage of change flattens out from positive to zero. These inflection points are also marked in Figure 2 for comparison.
|
Thirty-six percent of jobs advertised from January 1999 through December 2001 were for academic positions. This result is statistically unchanged from the previous 48-month period between January 1995 and December 1998 but is down from 47% during the period from January 1991 through December 1994. A recent widening of the gap between academic and private practice positions, which were very close from 1991 until the nadir of the market in 1995, has also occurred (Fig. 4).
|
Several geographic trends are noted when the data are compared with the results from our previous studies [9, 11] (Table 1 and Figs. 5,6,7). There has been a statistically significant increase in the relative number of jobs advertised in the Midwest, which accounted for 29% of total ads from January 1999 to December 2001 (up from 16% and 22% in the earlier and more recent previous 48-month periods, respectively). The proportion of advertised positions in the Northeast has decreased slightly but significantly (to 23%, from 27%, in both of the earlier time periods, respectively), as has the proportion of ads in the Southwest (to 8%, from 12% and 10% in the previous two periods, respectively). Meanwhile, the proportion of jobs in California has remained statistically unchanged at 6% for the period between January 1999 and December 2001 (after dropping from 8% to 6% across the previous two periods, respectively). This proportion is only 5%, though, over the most recent 24 months, representing a statistically significant decrease. The relative numbers of positions in the Southeast and Northwest have remained statistically unchanged from previous periods (at 27% and 6%, respectively).
|
|
|
The absolute number of ads, meanwhile, peaked in 2001 for the Northeast, Southeast, and Midwest but fell in 2001 versus 2000 for the Southwest (475 vs 540), Northwest (299 vs 334), and California (270 vs 321). Advertisements for California actually peaked in 1999 at 338 (Fig. 8).
|
Several subspecialties, including mammography, pediatric radiology, neuroradiology, abdominal imaging, and chest radiology, have all displayed statistically significant relative increases in demand for the period from January 1999 to December 2001, as compared with the previous two periods. Mammography accounted for 10% of the ads (up from 5% and 8%); pediatric radiology, for 5% (up from 4% and 3%); neuroradiology, for 10% (up from 8% and 8%); body imaging, for 15% (up from 10% and 14%); and chest radiology, for 3% (up from 2% during the previous 48-month period). On the other hand, statistically significant proportional decreases in demand have been observed in general radiology (down to 34%, from 43% and 41%) and vascular and interventional radiology (down to 13%, from 16%). There were no statistically significant changes in the proportions for musculoskeletal radiology and nuclear medicine. The changes in emergency radiology positions were statistically significant but small (Table 2 and Figs. 9,10,11).
|
|
|
Ads for MR imaging and sonography, which were included in the abdominal imaging category, accounted for 4.5% and 0.7% of the total positions, respectively, listed from January 2000 through December 2001. The full-time nighthawk positions included only 0.7% of the total from June to December 2001 and were coded within various subspecialty categories. As a result, their inclusion does not appear to alter the data analysis appreciably.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although confirming the existence of a severe radiologist shortage, our study has found that the improvement in the overall job market for diagnostic radiology, which began in December 1995 (Figs. 2 and 3), may have stabilized. The rate of job increase nationwide has slowed dramatically since early 2000 and appears to be reversing itself as of March 2001 (Figs. 2 and 3). This slowdown has been most pronounced in the western part of the country because the absolute number of positions advertised for the Southwest, Northwest, and California actually declined in 2001. Advertisements for California have been declining for over 2 years, indicating the market there may be becoming saturated. However, whether the nationwide market peak has indeed been reached is not clear, so continued advertisement surveillance over the next several months will be needed to resolve this issue.
These findings do not suggest that the market situation is reverting to the tough times of the mid 1990s. Instead, the current strong market appears to be continuing unabated but not becoming more overheated. Stabilization of the shortage is in fact desirable because being near equilibrium should prevent a major, detrimental undersupply of radiologists from developing, especially in California, the Northwest, and the Southwest. Currently, more than 50% of radiologists already report that they are overworked [17]; if the shortage continues to grow, as it has in recent years, the situation will become even more serious.
Because the number of entrants into the field has been fixed at a constant level (constrained by the number of residency graduates each year) and because imaging volume (demand) has increased substantially [18], this slowdown, if real, only can be accounted for by increases in productivity or increases in nontraditional physician supply. Indeed, technologic innovations, PACS (picture archiving and communication system) implementation, computer-aided diagnosis, and teleradiology operations may be improving productivity. In addition, there have been numerous informal reports of recruitment of retired and part-time radiologists to assist those overburdened by increased workload requirements. One cannot rule out, though, that radiology practices may have given up on journal advertising because the shortage has worsened and journal advertising could represent a declining method of finding radiologists. No evidence currently exists to substantiate or disprove this claim, but information analyzed from the American College of Radiology Professional Bureau (whose methods are relatively impervious to the problem of changing employment marketing techniques) has shown that the diagnostic radiology job market was actually more competitive in 2001 than in 2000 (Sunshine JH, unpublished data). Thus, changing recruitment practices, such as electronic advertising and the use of "headhunter" services, may have also helped account for the lower overall advertisement totals during the past few months without reflecting true changes in market demand.
Despite these influences, almost three times as many positions exist nationwide as there were 4 years ago. For every advertised position lost in California, the Northwest, and the Southwest, more vacant positions have been gained in the Midwest and Southeast. Market growth remains strongest in the Midwest, where the absolute number of advertisements per year has more than quadrupled and the relative proportion has almost doubled since 1995. In fact, it is not rare for many private and academic hospitals in the Midwest to be advertising to re-staff almost entire departments. This trend correlates well with the recent American College of Radiology report that nearly 600 radiology department chair, director, and section leader positions are currently vacant nationwide [19]. This lack of leadership is distressing, and this situation does not bode well for reversing the shortage of specialty physicians in more rural areas, such as those in the Midwest.
Changes in subspecialty demand have been less dramatic. The relative demand for general radiologists has continued to decrease, whereas the most pronounced relative increases in subspecialty demand have been witnessed in mammography and abdominal imaging, driven by growth for MR imagers. These increases in demand were most pronounced in California and the Northeast, where most of all advertisements in these regions for 2000-2001 were for these two subspecialties. Although increasing numbers of advertisements may suggest that radiologist shortages exist within these fields, the increase in numbers may alternatively indicate overall growth of those fields, with more positions being advertised to larger pools of fellowship-trained radiologists. Meanwhile, decreases seen in emergency radiology demand may be due to staffing of emergency departments with radiologists from other subspecialties, especially abdominal imagers.
Future Considerations
Although we believe that our help wanted index analysis has accurately
portrayed the current market situation, some caution should be exhibited when
using help wanted index data to predict long-term future trends. Help wanted
index forecasting is not immune from potential technology and policy changes
that have caused radiology job-market prediction to be so difficult
historically. Novel imaging techniques (such as functional MR imaging and
molecular and and genetic imaging) and new interventional techniques may
substantially increase demand for radiologists in the next several years in
ways that cannot be calculated, just as new technologies have in the past. For
example, in 1990, over 60% of radiology work was derived from techniques like
CT, MR imaging, and sonography that were not available only 25 years before
[4]. On the other hand,
productivity gains because of PACS, computer-aided diagnosis, and
teleradiology may help to offset some of the increased volume. In addition,
health care reform and changes in medical delivery systems can potentially
impact volume of imaging demand, thus altering market conditions. In short,
the help wanted index projects tomorrow's market conditions on the basis of
today's realities. If these assumptions are altered, then the future becomes
less predictable.
Regardless of these limitations, in an ideal situation supply and demand would be nearly matched. In the future, if demand continues to outpace supply, one of three eventualities can be expected: radiology as a specialty will not continue to provide optimal patient care; patients will be diverted to other subspecialists not trained in imaging; or demand on individual radiologists will exceed their ability to provide optimal interpretation. None of these three options is desirable.
One example of what can happen because of a substantial shortage in supply of trained radiologists is seen today in vascular and interventional radiology. With improved techniques for minimally invasive diagnostic and therapeutic procedures, the job market for vascular and interventional radiology has increased dramatically during the past decade. In fact, from the nadir of 74 ads in 1994, the number of positions advertised has increased over 10-fold to 771 ads in 2001. At the same time, the number of interventional radiology fellowship positions has remained relatively stable at 200 nationwide. Clearly, we are not training enough to meet the increasing demand.
Although this news may be good for residents and fellows searching for jobs, other medical specialists have stepped in to fill the void by performing interventional procedures themselves. In comparison with the 200 interventional radiology fellows, 750 invasive cardiologists graduate and enter the workforce each year. Today, many cardiology practices are repositioning themselves as cardiovascular ones, interpreting cardiac and peripheral MR angiograms and performing carotid, renal, and iliac artery stenting procedures [20]. Almost 20% of invasive cardiology training programs teach peripheral vascular disease management. Recognizing this trend, the American College of Radiology recently approved a resolution to recognize vascular and interventional radiology as a separate component of radiology. The Society of Cardiovascular and Interventional Radiology supported this resolution and hopes to increase the number of practicing full-time equivalents in vascular and interventional radiology by 40% before 2006 [21].
Interventional radiologists are by no means the only group affected by the current shortage. The number of radiology residency positions nationwide has decreased in the past few years because of anticipated funding restrictions as a result of the Balanced Budget Act of 1997, decreasing the number of entrants into all subspecialties. Emergency department physicians, orthopedic surgeons, obstetricians, nephrologists, and cardiologists (to name but a few) have responded to the shortage of appropriately trained radiologists and now interpret images or perform imaging-guided procedures themselves to expedite patient care. Many radiology groups have countered by increasing staffing in emergency departments, either with 24-hr in-house coverage or from home through teleradiology to limit their loss of market share.
Potential Solutions
It is of paramount importance that our national organizations respond to
this crisis and work to increase the number and efficiency of radiologists in
tomorrow's workforce. From our study, the most critical shortages currently
exist in the fields of mammography and pediatric and abdominal imaging, so
potential solutions should be crafted with their needs in mind.
A short-term solution would be to increase the number of part-time, retired, and internationally trained radiologists in the workforce until a long-term solution can be implemented. Sunshine et al. [17] have documented that currently 10% of posttraining radiologists are working part-time in radiology, and 17% are retired and are not working at all. Although it may be difficult to succeed, efforts to tap into this labor pool could only help. Also, general radiologists (for whom demand is declining) could be encouraged, through possible government or private radiology-group subsidies, to undergo additional training in mammography and pediatric or abdominal imaging, allowing them to perform more of the work.
As for the long-term, the Harkin-Snowe Bill, introduced in the 107th Congress and sitting in committee at the time of this writing, would provide more Medicare funding for additional radiology residency positions. This legislation could therefore potentially alleviate the shortage of radiologists by boosting supply in the next several years. This bill is not necessary, though, because the Center for Medicare and Medicaid Services does not restrict the number of trainees, only the funding. Radiology groups could fund residency slots in return for guarantees that residents join their practices after they graduate, like corporations do for Masters of Business Administration students.
In addition to workforce increases, a larger potential source of benefit may reside in increasing radiologist productivity. Although training 5-10% more radiology residents would be difficult and expensive, it may be easier to obtain 5-10% greater efficiency from radiologists currently in practice without appreciably increasing workload. We are all aware of the amount of time spent on noncognitive issues; we often perform clerical functions that are more efficiently performed by others. We also have not fully captured the efficiencies of new technologies and instead have incorporated them into old processes rather than rethinking our entire way of providing our services in the setting of technologic innovation. With modest capital expenditures and a willingness to accept change, many radiology practices (that have not already done so) could begin to implement Internet—based PACS, teleradiology, radiology information systems, and voice recognition systems. Although, to our knowledge, no studies in the current literature substantiate the benefits of these technologies, scattered reports indicate promise. Our practice settings must be enabled to allow this potential benefit. Merely adding voice recognition systems, when dictation was previously the rule, does not improve efficiency. Only with a shift in the previous resources to better use this technology, can it be thought of as helpful.
These technologies could not only expedite information processing but also increase coordination between different radiology centers. Increasing flexibility could make it easier for radiology groups to outsource work to others in times of peak demand and, thus, distribute work more effectively through pooling resources. Specifically for mammography, development of digital mammography and computer-aided diagnosis systems could eliminate the need for a second interpretation, thus decreasing the workload substantially, although we recognize that these systems have not yet been perfected. Meanwhile, in the clinical setting, increasing reliance on allied health care professionals, such as nurse practitioners, clinical nurse specialists, and physician assistants, for patient-related tasks would free up radiologists to perform more technically complex and demanding duties.
Finally, redistributing radiology trainees could serve as another means of alleviating the shortage. Residency program directors and advisors should understand the market situation and guide trainees into areas of increasing need, including mammography and pediatric and abdominal imaging. As Anzilloti et al. [10] showed, physicians respond to market demands: shortages lead to increased salaries, and increased salaries lead to increased entry. Fellowship cycles (1-2 years) are much shorter than residency cycles (4-5 years), so mismatches across different radiology subspecialties can be addressed in a more efficient manner than the shortage in radiology as a whole.
The primary goal of increasing the number of radiologists to meet demand is to provide optimal patient care. This goal includes several factors: having enough radiologists to perform and interpret imaging studies, providing time and resources to develop new imaging and interventional techniques, and minimizing diversion of studies to other physicians who are not trained to perform or interpret imaging studies. Further, we must have a sufficient number of educators to train the next generation. Therefore, it is not only in our own interest, but also in our patients' best interests for us to understand the job market to be in a position to protect our "turf."
The current shortage of radiologists entering the workforce requires immediate attention for crafting a long-term solution. For the well-being and the future of our specialty, we will need to supply an adequate number of well-trained radiologists to provide optimal patient care, to fight and win turf battles, and to allow us to expand with advances in imaging technology. In other words, we must as a group act now to preserve our specialty in the future.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Bhargavan, A. H. Kaye, H. P. Forman, and J. H. Sunshine Workload of Radiologists in United States in 2006-2007 and Trends Since 1991-1992 Radiology, August 1, 2009; 252(2): 458 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Licurse, D. D. Saket, J. H. Sunshine, C. D. Maynard, and H. P. Forman Update on the diagnostic radiology employment market: findings through 2005. Am. J. Roentgenol., September 1, 2006; 187(3): W249 - W254. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. Saket, C. C. Nwanze, C. D. Maynard, J. H. Sunshine, and H. P. Forman Update on the Diagnostic Radiologist Employment Market: Findings Through 2004 Am. J. Roentgenol., December 1, 2005; 185(6): 1408 - 1415. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. Saketkhoo, M. Bhargavan, J. H. Sunshine, and H. P. Forman Emergency Department Image Interpretation Services at Private Community Hospitals Radiology, April 1, 2004; 231(1): 190 - 197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine, C. D. Maynard, J. Paros, and H. P. Forman Update on the Diagnostic Radiologist Shortage Am. J. Roentgenol., February 1, 2004; 182(2): 301 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. Saketkhoo, J. H. Sunshine, A. M. Covey, and H. P. Forman Findings in 2002 from a Help Wanted Index of Job Advertisements: Is the Job-Market Shortage of Diagnostic Radiologists Easing? Am. J. Roentgenol., August 1, 2003; 181(2): 351 - 357. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
