| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Research Department, American College of Radiology, 1891 Preston White Dr.,
Reston, VA 20191.
2 Committee on Radiologist Resources, American College of Radiology, Reston, VA
20191.
3 Department of Diagnostic Imaging, Brown University Medical School, Box G,
Providence, RI 02912.
4 Department of Diagnostic Imaging, Rhode Island Hospital, 593 Eddy St.,
Providence, RI 02903.
Received October 2, 2001;
accepted after revision November 15, 2001.
Address correspondence to M. Bhargavan.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We used responses from the American College of Radiology's 2000 Survey of Diagnostic Radiologists and Radiation Oncologists to determine the current age distribution and activity of diagnostic radiologists. The numbers entering the profession were projected using three assumptions: no change in training programs, reduction of residency to 3 years (or otherwise increasing the annual number of graduates by one third), and elimination of most fellowship programs. Demand projections assume a 5% shortage in 2001 and depend on growth rates of the population, aging, scenarios of growth of age-standardized demand, and the effect of possibly productivity-enhancing technologies such as PACS (picture archiving and communication systems).
RESULTS. Only a one-third increase in annual graduates materially increases the work-force relative to current training patterns. In all cases, the growth rate of the demand for radiologists far outstrips the supply over a 30-year horizon. In the shorter term, projections of current trends point to an increasing shortage, but rapid major productivity increases could produce a surplus.
CONCLUSION. Those in the field of diagnostic radiology should consider measures to mitigate the increasing shortage, while monitoring developments that might signal departures from current trends in supply and demand.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Accurate projections of demand and supply, to the extent that they are possible, are useful. A series of articles over the years made [3,4,5,6,7,8,9] and examined [10] these predictions for radiologists. The ACR conducted several surveys to track trends in hiring, job search, job satisfaction, and perceptions of their employment market to help diagnostic radiologists understand the employment market in the near past and near future and to use this information to correct long-term imbalances, if any [1, 2, 11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33].
Projections of medium and long-term manpower needs and supply in diagnostic radiology help the profession determine whether policy action is warranted and if it is, the extent of such an action. For example, although the profession currently faces a shortage, correcting it would not be productive if the shortage is likely to end in the next few years. Projections over a long time horizon can help assess whether the current shortage is merely a short-term fluctuation or is a long-term imbalance, the latter being a greater cause for concern and action.
Accurate projections of surplus and shortage are difficult because of the multitude of uncertain factors that cause and influence these imbalances. However, despite the difficulty, a modeling exercise is not useful unless it takes these factors into account. The supply of radiologists is affected by the number of trainees entering the profession and by the rate at which practicing radiologists retire. Demand for radiologists is affected by the growth rate and aging of the population, the growth rate of demand per age-adjusted American, and the number of procedures that can be performed by each radiologist given the current state of the technology. Of these, growth in age-adjusted demand and changes in technology affecting the workload that a radiologist can handle are the most difficult to predict.
Using a time horizon that stretches from the present to 2030, we compared three projections of the supply of radiologists with several plausible scenarios for demand. With regard to supply, in addition to the current pattern of training, we considered two options for changing training: either reducing radiology residency from 4 to 3 years or eliminating postresidency fellowships for all the less essential subspecialties. These supply options were considered by the Human Resources Task Force assembled by the ACR in 2001 from radiology leaders of many organizations to consider actions related to the shortage. For each of the two changed training options and the current training pattern, we traced the number of active posttraining full-time equivalent (FTE) radiologists over the next 30 years and compared these supply projections with several possible scenarios for demand. The demand scenarios differ in the hypothesized rates of growth in demand based on various historical values of the rate of growth of age-standardized demand and differ in estimates of the effect of productivity-enhancing technologic changes. The productivity-enhancement patterns that we modeled were based on guidance from the task force.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Medicare restricts the number of available residency slots. As a result, the number of radiologists completing training is expected to be a fairly stable number, close to the current level, if the status quo in training programs is maintained. The total number of residents in 2000 and 2001 was obtained from ACR's survey of directors of accredited diagnostic radiology training programs in the United States (Hendershot M, personal communication). (The response rate of 98% makes the survey results fairly reliable.) This number could be changed by policy modifications such as redesigning the training program. We modeled two possible alternatives to the current system of training that resulted in an increase in the number of radiologists entering the profession. The first alternative that we considered was that the length of residency in radiology would decrease from 4 to 3 years. Because we assume that Medicare regulation keeps the total number of residency slots fixed, this change would increase the annual number of residency graduates by one third. The second alternative that we modeled was that the fourth year of radiology residency would be formalized as a "fellowship" year for all subspecialties except for neuroradiology and vascular and interventional radiology, in which the postresidency fellowship year is deemed most critical. Currently, approximately 75% of graduating residents go on to 1-year or longer fellowships and enter the workforce subsequently. In our second alternative, apart from the two stated fields, no one advances to a postresidency fellowship. This plan leaves 30% of residents going on to a fellowship. (To keep the model simple, we assumed that all fellowships are 1 year long. In the model, new radiologists thus enter the workforce either directly after the completion of their residency or 1 year later after completion of their fellowship. The specific fellowships retained in the model were suggested by the ACR-assembled task force, but the results are essentially identical for any situation in which most residents cease taking postresidency fellowships.)
From the ACR's 2000 Survey of Diagnostic Radiologists and Radiation Oncologists [33], we obtained information on radiologists' age distribution, the percentage in each age category retired, and the percentage in each category working part-time. The surveyed sample included residents, fellows, active posttraining physicians, and retirees. The respondents in 2000 were similar to the respondents in a 1995 survey in terms of the percentage retired in each age category and the percentage working part-time in each age category [19, 33]. Therefore, we assumed that current values for these two percentages will persist over the next 30 years. Each part-timer's work was assumed to be 70% of that of a full-time radiologist. This conversion factor was based on data on average work hours from the ACR's 1995 Survey of Radiologists [19].
Death rates by age were based on survivorship statistics obtained from the National Vital Statistics Report [34]. White men in the United States in the highest income and education category have a life expectancy that exceeds the population average by 2 to 3 years [35]. We used the finding of Sunshine et al. [33], the difference in mortality rates between men and women and the number of women in radiology, to estimate that the age-specific death rate of radiologists is approximately 70% of that of white men in general. We assumed that people will live longer in the coming years. Specifically, we modeled death rates as decreasing at the same rates as they did over the past few decades.
Methodology for Supply Projections
Responses from the 2000 Survey of Diagnostic Radiologists and Radiation
Oncologists [33] were weighted
to mimic the entire population of diagnostic radiologists in 2000 as a
starting point. We used the ages reported in the survey in combination with
assumptions about the number and age-distribution of new physicians entering
the profession yearly to calculate the age distribution of active physicians
in each year and aged our model 1 year at a time. We assumed that those
completing residency had the same age distribution as that of the combination
of residents and fellows in the sample. When assuming that the current
training program system continues, we modeled 75% of those who completed their
residency last year (and went on to a 1-year fellowship) and 25% of those
completing their residency this year as entering the workforce. For the 3-year
residency option, we assumed that completing residents were 1 year younger at
the end of the residency program. In the option in which most fellowships are
eliminated, more of the completing residents would directly enter the
workforce, as opposed to entering the workforce in the following year. Both of
the training program alternatives that we considered resulted in a slightly
younger mix of new radiologists relative to the current training program.
We applied age-specific mortality rates (a separate rate for each 1-year age group) to the number of radiologists of each age to determine how many physicians in each age category survived each year. We applied the currently observed (year 2000) percentage of those working full-time and part-time in each age category to the calculated future age distributions and, thereby, obtained predictions for the numbers of full-time and part-time radiologists. Applying the conversion factor of 0.7 to each part-time physician, we calculated the number of FTE physicians each year.
In the two training program changes that we modeled—namely, reducing the length of the residency or largely eliminating the postresidency fellowship year—we assumed that the change would take effect for the class that graduates in 2006. Changes in requirements would, thus, have to take effect in 2002 or 2003.
Demand Assumptions
Current demand was assumed to be 5% larger than supply, where supply is
measured in FTEs. This assumption was based on the data about the shortage
already cited and the finding from the 2000 survey that 51% of diagnostic
radiologists reported they had too much work, whereas only 5% said they had
too little work [33].
The growth rate of the population in the United States and the increase in demand resulting from population aging were taken from the Office of the Actuary, Health Care Financing Administration (now the Centers for Medicare and Medicaid Services) [36, 37]. Population was projected to grow at 0.83% per year currently and steadily slower thereafter, with growth at 0.46% annually by 2030. The annual demand increase resulting from aging was 0.34% in 2001 and is expected grow to 0.78% in 2030.
The rate of growth of demand per age-adjusted American was calculated from historical data, using the Medicare Physician/Supplier Procedure Summary data file (which was formerly known as Part B Medicare Annual Data [MBAD] Procedure file). We used the average growth rate of procedures and relative value units (RVUs) per age- and disability-status standardized beneficiary. The growth rate of procedures was an underestimate of growth in the demand for radiologists because more recently developed procedures, in particular interventional radiology, MR imaging, and CT, are growing more rapidly than other procedures and are more complicated and require more radiologist time per procedure than earlier ones. On the other hand, the growth rate of RVUs exaggerates the growth in workload because RVUs per hour are higher with high-tech procedures [38, 39]. Our methodology assumes that after adjusting for changes in the age structure of the population, radiology use in the elderly and nonelderly would increase at the same rate—say 3% annually. Of course, the level (as opposed to the rate of increase) in the elderly is much higher, approximately four times as high, on average.
An important element that affects the number of radiologists needed is the number of procedures that each radiologist can perform in a given period (say 1 hour or 1 year). (This number will obviously vary widely across modalities.) Recent technologic developments such as PACS (picture archiving and communication systems) may have a significant impact on radiologists' productivity. The literature on PACS [40,41,42,43,44] is not helpful in answering the question of what to expect. In general, it indicates that PACS save radiologists and technologists time. Despite more preparation work, the easy access to images and the voice-recognition software that often accompanies PACS result in less time needed to generate reports. However, the literature concentrates on report turn-around time, whereas our concern was productivity—the number of examinations that can be completed in a fixed-length workday. Also, most research reports on a single modality, not a full department, and "publication bias"—that is, the tendency to report successes and not failures—is probably strong. Computer-aided detection and diagnosis is another technology, currently in an even earlier state of development, that may substantially increase productivity.
Methodology for Demand Projections
We explored five scenarios of demand for radiologists. These combined three
different rates of growth of per capita age-adjusted demand for radiology with
three possible impacts of time-saving applications (like PACS) on the number
of radiologists required.
We assumed a conservative baseline scenario, which we called the "Main Scenario." The annual increase in demand per age-adjusted person, otherwise known as "technologic progress," is assumed to be 1.75%. This is the average of the growth in procedures (1%) and growth in RVUs (2.5%) per age-adjusted Medicare beneficiary between 1992 and 1998. Adding to this the rate of growth of population and the increase in demand as a result of the aging of the population, we arrived at a total annual rate of growth of 3-3.1% for the Main Scenario.
We considered four other scenarios, ranging from a collapse in demand at one extreme to fast-growing demand at the other extreme.
The second scenario uses the same growth rate in demand for procedures per age-adjusted person as the main scenario (i.e., 1.75%) but assumes that productivity-enhancing changes in how radiologists work—changes such as PACS—reduce the demand for radiologists. We called this scenario "Main with PACS." As a result of productivity enhancements, demand for radiologists is assumed to decrease by 20% over the next 10 years (i.e., by 2011). Twenty percent for the decrease is loosely based on the literature on PACS [40,41,42,43,44]. After 2011, productivity-enhancing changes are assumed to be fully disseminated with no more productivity gains. Therefore, after 2011, the growth rate in demand is the same as in the Main Scenario. In this scenario, the total annual rate of growth of demand over the next 10 years is 0.7-0.8%; thereafter, it is 3.1%. (Note that because multiplication of percentages and addition of percentages differ, the rate for the first 10 years is <1%, not 1.1%.) The compound average annual rate of growth of demand over the entire period from 2001 to 2030 is 2.3%.
The third scenario that we considered was "Faster Growth with PACS." We assumed that age-adjusted demand grows at the rate of 3.75%, which is the average of the 3% growth in demand for procedures (as observed in the Medicare Physician/Supplier Procedure Summary data during the period from 1986 through 1998, the full period for which data were available to us), and the 4.5% growth in RVUs during the same period. The effect of productivity-enhancing changes such as PACS was assumed to be the same as that in the previous case: it decreases demand for radiologists by 20% over the next 10 years. The total annual rate of growth of demand, including the effect of population growth and aging, is just under 3% for the next 10 years and over 5% thereafter. The compound average rate of growth from 2001 to 2030 is 4.3%.
The fourth of our scenarios is called "Demand Collapse." Here, we assumed an annual increase in demand per age-adjusted person of only 0.5%, approximately that seen in the Medicare population in 1992-1995. We followed this scenario for only 10 years, because it seemed highly improbable that even a tight regime of controls could keep age-adjusted demand growth to only 0.5% for even a decade. We assumed a higher productivity gain from productivity-enhancing changes than that in the previous scenarios, with these changes decreasing demand by 30% over the next 10 years. The total annual rate of growth of demand over the next 10 years is-1.8%. This scenario was deliberately designed to be an extreme case and was not shown in the graph (Fig. 1).
|
The last scenario was also designed as an extreme one. In it, age-adjusted demand per capita grows at 3.75%, the rate at which it did in 1988-1998 and the rate assumed in Faster Growth with PACS. However, we assumed that productivity-enhancing changes either do not become widely prevalent or if new technologies like PACS become prevalent, they do not deliver any net productivity gains for radiologists. We called this scenario "Faster Growth with No PACS." In this case, the total annual rate of growth of demand would be 5.1%. (Again, this scenario is not in the graph [Fig. 1].)
Starting from a demand for radiologists in 2001 that is 5% higher than the supply measured in FTE, we calculate the numbers of FTE radiologists that will be needed over the next 30 years under each of the five demand scenarios. We combine these projections with the three supply projections—each of which represents an option for training program structure—to calculate surpluses or shortages for each year for each combination of demand scenario and training program option.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Table 1 contains the numerical supply projections underlying Figures 1 and 2. It contains the projected number and FTE of radiologists at approximately 5-year intervals beginning in 2001 and continuing through 2030. Table 2 indicates the difference in the number of radiologists available relative to the current training system, if either change in training programs is implemented. Basically, the option of shortening residency to 3 years would increase the number of radiologists entering the workforce by one third each year. As a result, the FTE of the radiologist workforce would be 7% larger by 2010, 16% larger by 2020, and 24% larger by 2020 (Table 2) than that with current training arrangements.
|
|
Applied literally, the 3-year residency results in a lumpy growth in the number of physicians entering radiology. With a fixed number of residency slots, two sets of residents will complete training in 2006—namely, the last class of 4-year graduates and the first class of 3-year graduates. These will be replaced by an entering class of residents that is twice as large as that in the previous normal year. These will again graduate and be replaced by an entering class of residents 3 years later, resulting in 3-year cycles with one class twice as large as the other two. Designers of training programs presumably would want to smooth out this lumpiness in some way.
Option 2, fellowships replaced by the fourth year of residency for most subspecialties (all except neuroradiology and vascular and interventional radiology), results in a one-time increase of the workforce by approximately 480 radiologists in 2006 relative to numbers under current arrangements. In that year, entrants to the posttraining workforce would consist of both a large number of fellowship graduates from the last class in which most radiologists did fellowships and a large number of residency graduates from the first class in which most residents do not go on to a fellowship. However, after that year, option 2 does not increase the number of radiologists entering the workforce each year beyond the number produced by current training arrangements. Thus, the effect of option 2 on the FTE of the workforce is a 1.7% increase in 2006 beyond what current arrangements will produce, but thereafter, the effect, in percentage terms, dwindles gradually (as the total FTE becomes larger and as some of the extra graduates of 2006 retire), falling to 1.4% by 2030.
Table 3 and Figure 1 show the projected differences between supply and demand for radiologists under the different options for training programs and the different scenarios for demand.
|
In the Main Scenario, there is a shortage of radiologists throughout the period between 2001 and 2030. The shortage grows to more than 90% in 2030 with a 4-year residency, with or without a formal fellowship year. The 3-year residency keeps the shortage down to only about 56% in 2030.
If demand grows at a conservative rate but productivity-enhancing changes in radiologists' work increase radiologists' productivity (Main with PACS), then there is still the possibility of a surplus during the first decade studied, but the growth rate of demand overtakes the growth of FTE radiologists after that period. With this demand scenario, if the residency is shortened, then demand overtakes supply later in the second decade relative to the other options.
With the higher rate of growth in demand in Faster Growth with PACS, with the same impact of productivity-enhancing changes as in Main with PACS, there is a shortage of 120-180% by 2030, depending on the chosen training program option.
With each of the options for the residency program, there is a surplus of radiologists under the Demand Collapse scenario. This scenario is only extended to 10 years, and by 2011, there is a surplus of 29% with a 3-year residency, 24% with no change in training programs, and 25% with fewer fellowships.
In the case of fast growth in demand with no impact of productivity-enhancing changes, the shortage in the long run could be as large as 250% if there is no change in the training program and close to 180% even with a 3-year residency.
In all scenarios followed to 2030, if technologic changes or changes in work organization fail to provide any major increase in radiologists' productivity or if the growth rate of demand for procedures per age-adjusted person matches the long-term trend rather than the slower Medicare rate for 1992-1998, then a persistent shortage of radiologists seems inevitable in the long run.
Table 4 shows the compound annual average rates of growth of demand and supply from 2001 and beyond for each of the supply options and demand scenarios. If training patterns remain unchanged, the annual rate of growth of the workforce (measured in FTEs) is 1.6% between 2001 and 2006 and only 1% over the full period from 2001 to 2030. With a 3-year residency, the growth rate of FTE radiologists is higher, starting at close to 2% per year and equaling 1.7% per year over the entire 2001-2030 period. The option of replacing most fellowships with a specialization in the fourth year of radiology results in a one-time increase of approximately 1.9% in radiologists in the work force in the year that most fellowships are eliminated, but subsequently, the growth of the workforce is at the same rate as that with existing training programs.
|
Demand for radiologists grows at 2-5% in the long run, depending on which demand scenario one considers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Some would consider our projections of the growth of demand extremely conservative. The highest rate of growth of procedures per age-adjusted American that we model is 3% annually, the long-term Medicare rate. In contrast, one private insurer (MedSolutions) that covers approximately 127 million member months of commercial-aged outpatients (Poenitske A, personal communication, 2001) reports a rate in excess of 7% annually from 1997 to 2000. If demand grows much more rapidly than we have modeled in our report, then the shortage of radiologists will, obviously, be much greater.
With respect to training program changes, doing away with most fellowships does not significantly alleviate the anticipated shortage of radiologists in the long run, or even in the short run. Only a switch to a 3-year residency significantly affects it.
The same effect as a switch to a 3-year residency would be achieved by increasing the number of residency slots for radiology by a third and keeping residency at 4 years. This change would likely encounter major opposition from other specialties not wanting to give up slots or from the Centers for Medicare and Medicaid Services (CMS, formerly Health Care Financing Administration), not wanting to increase costs by increasing the total number of slots it funds. On the other hand, the idea of shortening residency to 3 years raises serious questions of the adequacy of training that can be provided in only 3 years, and for this reason, the Human Resources Task Force did not recommend it.
Training institutions may create slots that are not funded by CMS but have rarely done so because of the lack of funding. Radiology training programs might look to private radiology groups to fund some slots, in exchange for the trainee's agreement to work for the group. This plan is another way to increase the annual number of graduates. However, some radiology residency programs might have difficulty providing the additional faculty required by any increase in residents.
Eliminating the now required clinical year of radiology training would not increase the annual number of graduates except in the instances in which the radiology department "owns" the clinical-year slots and thus could redeploy them to the radiology residency per se. Because this situation is rare, eliminating the clinical year would have a negligible effect.
Although our projections generally point to a shortage, this outcome is not assured. Some of our scenarios show an approximate balance or even a large surplus in 10 years. Moreover, the history of predictions of physician—or radiologist—surplus or shortage is filled mostly with failed predictions. Most memorably, the Graduate Medical Education Advisory Committee (GMENAC) predicted a surplus of radiologists by 1990, with that surplus growing even larger by 2000. In fact, the number of radiologists in practice in 1990 was 15% less than the GMENAC had predicted [11], and 2000 was characterized by a shortage.
The effect of uncertainties in predictions is multiplied by the fact that the annual number of graduates is small relative to the total radiologist workforce. With roughly 1,000 graduates a year and 25,000 radiologists in the work-force, practices in any year revising their staffing needs by 1% constitutes a change equal to 25% of the year's graduates.
If productivity-increasing changes are much larger than those that we have modeled, they could possibly meet the shortage or lead to a surplus. Productivity-enhancing possibilities, the effect of which we have not attempted to specifically quantify, include more efficient off-hours coverage using teleradiology, with radiologists working full-time to cover off-hour shifts in another region of the country or world (these radiologists might be located where the shift is daytime locally) and use of higher level non-physician staff for some noninterpretive tasks now performed by radiologists.
Finally, some self-correcting mechanisms may diminish—or even eliminate—any surplus or shortage. For example, some reductions in radiology training slots planned by teaching hospitals in the mid 1990s, when the radiology market appeared soft, were subsequently cancelled when the shortage appeared. Similarly, we assumed that retirement rates do not change in response to changes in demand for radiologists, but a persistent shortage of radiologists could lead to increased remuneration, which, in turn, could induce radiologists to retire later. Also, if radiologists continue to feel overworked, they may want to give up some least desired "turf" to others, thereby decreasing demand for their services. Alternatively, if the shortage of radiologists and technologists makes radiologists unable to provide some services desired by treating physicians, those physicians may take some of this turf. An issue here is whether radiologists could cede the turf they least want or whether treating physicians would instead succeed in taking on only relatively desired turf.
Acknowledgments
We thank James Andrews and Alan Poenitske of MedSolutions for drawing our
attention to the existence of a study on the growth of demand for radiology
procedures among commercial-age patients and for sharing their results with
us. We also thank Nancy Ellerbroek from the ACR Committee on Radiologist
Resources for her helpful comments. We thank the groups that responded to our
survey; their response makes information available to the entire
profession.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. M. Colletti The Diagnostic Radiology Exam of the Future: The Law of Unintended Consequences Meets the Law of Supply and Demand Am. J. Roentgenol., May 1, 2008; 190(5): 1147 - 1148. [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Wright, J. B Killion, J. Johnston, V. Sanders, T. Wallace, R. Jackson, D. Snodgrass, and R. A Redd RAs Increase Productivity Radiol. Technol., March 1, 2008; 79(4): 365 - 370. [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Bowen, W. E. Torres, and W. C. Small Nonphysician Providers in Radiology: The Emory University Experience Radiology, October 1, 2007; 245(1): 3 - 6. [Full Text] [PDF]
|
|
|
|
|
|
C. Meghea and J. H. Sunshine Retirement patterns and plans of radiologists. Am. J. Roentgenol., December 1, 2006; 187(6): 1405 - 1411. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine, and C. Meghea, How could the radiologist shortage have eased? Am. J. Roentgenol., November 1, 2006; 187(5): 1160 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. I. Meghea and J. H. Sunshine Who's Overworked and Who's Underworked among Radiologists? An Update on the Radiologist Shortage Radiology, September 1, 2005; 236(3): 932 - 938. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chan and R. B. Gunderman Emerging Strategic Themes for Guiding Change in Academic Radiology Departments Radiology, August 1, 2005; 236(2): 430 - 440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. T. Bramson and G. A. Taylor SOS: Can We Save Pediatric Radiology? Radiology, June 1, 2005; 235(3): 719 - 722. [Full Text] [PDF]
|
|
|
|
|
|
M. Bhargavan and J. H. Sunshine Utilization of Radiology Services in the United States: Levels and Trends in Modalities, Regions, and Populations Radiology, March 1, 2005; 234(3): 824 - 832. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine and Y. S. Cypel Types of Procedures Performed by Diagnostic Radiology Practices: Past Patterns and Future Directions Am. J. Roentgenol., November 1, 2004; 183(5): 1193 - 1201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kalyanpur, V. P. Neklesa, D. T. Pham, H. P. Forman, S. T. Stein, and J. A. Brink Implementation of an International Teleradiology Staffing Model Radiology, August 1, 2004; 232(2): 415 - 419. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Smith, J. Temte, J. W. Beasley, and M. Mundt Radiographs in the Office: Is a Second Reading Always Needed? J Am Board Fam Med, July 1, 2004; 17(4): 256 - 263. [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]
|
|
|
|
|
|
Y. S. Cypel and J. H. Sunshine Basic Characteristics of Radiology Practices: Results from the American College of Radiology's 1999 Survey Am. J. Roentgenol., August 1, 2003; 181(2): 341 - 349. [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]
|
|
|
|
|
|
N. R. Dunnick Report of the 2002 Intersociety Commission Meeting: Radiology 2002--Today's Research Is Tomorrow's Practice Am. J. Roentgenol., April 1, 2003; 180(4): 925 - 928. [Full Text] [PDF]
|
|
|
|
|
|
S. R. Baker Abdominal CT Screening: Inflated Promises, Serious Concerns Am. J. Roentgenol., January 1, 2003; 180(1): 27 - 30. [Full Text] [PDF]
|
|
|
|
|
|
M. Bhargavan and J. H. Sunshine Workload of Radiologists in the United States in 1998-1999 and Trends Since 1995-1996 Am. J. Roentgenol., November 1, 2002; 179(5): 1123 - 1128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Grumbach Fighting Hand To Hand Over Physician Workforce Policy Health Aff., September 1, 2002; 21(5): 13 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine, R. S. Lewis, B. Schepps, and H. P. Forman Data from a Professional Society Placement Service as a Measure of the Employment Market for Physicians Radiology, July 1, 2002; 224(1): 193 - 198. [Abstract] [Full Text]
|
|
|
|
