| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Direct Research, LLC, 506 Moorefield Rd., S.W., Lower Level, Vienna, VA 22180
(chogan@directresearch.com).
2
Research Department, American College of Radiology, 1891 Preston White Dr.,
Reston, VA 20191 (jonathans@acr.org).
3
Committee on Radiologist Resources, American College of Radiology, Reston, VA
20191.
4
Department of Diagnostic Imaging, Brown University Medical School, Box G,
Providence, RI 02912.
5
Department of Diagnostic Imaging, Rhode Island Hospital, 593 Eddy St.,
Providence, RI 02902.
Received August 4, 2000;
accepted after revision August 14, 2000.
Address correspondence to J. H. Sunshine.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A survey was mailed to a sample of 970 radiology practices in the spring through fall of 1999; 73.0% responded. Responses were weighted to represent all practices in the United States providing diagnostic radiology services. Findings were compared with results of similar surveys from previous years.
RESULTS. In 1998, multiradiologist groups sought to hire 2299 (±148 [standard error]) diagnostic radiologists, an increase of 20% from 1977. Positions offered in 1998 were split almost equally between expansion positions and replacements for those who had left a practice. Another 422 (±58) positions had been vacated that practices did not seek to refill. In 1998, 621 (±57) diagnostic radiologists left active practice, which is more than the usual number of approximately 400 annually but less than the approximately 800 of 1996 and 1997. There was no significant association between hiring activity and the self-perceived effect of managed care on a group.
CONCLUSION. The pace of hiring and turnover of diagnostic radiologists increased in 1998. Positions available continue to exceed radiologists available to fill them; the excess was approximately 330 positions, which is not significantly different from the excess reported for 1997, but is higher than the (not statistically significant) net shortfall of 51 positions estimated for 1996.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The sixth such survey dealt with hiring in 1998 and was completed in the fall of 1999. The survey included questions pertaining to the number of positions for radiologists that groups sought to fill and the number of departing radiologists for whom groups did and did not seek a replacement. This article presents the survey findings.
We compare some core information with data from the 1997 survey [21] to study the change in the labor market since our last report. For consistency with the 1997 data, comparisons with the 1997 data include only multiradiologist groups and exclude single-radiologist practices—that is, solo practices.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The practices surveyed encompassed all types of physician practices at which radiologists work, including radiology units of multispecialty groups and government facilities, academic radiology groups, and private nonacademic radiology groups, and radiation oncology groups. In contrast to previous years, single-radiologist practices were included in this survey. Data for single-radiologist practices are tabulated separately and are not used in the analysis of trends in hiring.
Respondents indicated the type of practice by first answering if the group was primarily academic or nonacademic and then indicated which phrase best described the practice (government employees [i.e., Department of Veterans Affairs, military], part of a private multispecialty group of part of a private radiology group, or other). For this study, all self-reported academic practices are considered a single category, and all nonacademic practices are categorized by the self-reported description of the practice.
The sample of practices was drawn from a list of 2965 responses to a summer 1998 mailing by the ACR. The mailing asked ACR members to return a request form if they wanted to receive a copy of the ACR's "Compliance Program Guidelines for Radiologists and Radiation Oncologists" for their practice. The request form included information about the practice name, contact person, and mailing address. After removal of duplicate and incomplete requests, 2448 practices remained. The survey sample consists of 970 practices drawn at random from this list of 2448 practices and includes both single- and multiradiologist practices.
The survey encompassed five separate mailings; the first four were sent by regular mail and a final mailing was sent via priority mail. Handwritten notes were added to the cover letters for the third and fourth mailings, and the ACR and Radiology Business Managers Association leaders from the five most populous states sent e-mail reminders to nonrespondents in their states.
We received 708 responses to the survey. Of these responses, 30 were excluded for being out-of-scope (e.g., residents in training), and another 33 responses were excluded for substantially incomplete data—typically, for a lack of information about the number of radiologists in the practice. The overall response rate was 73.0%, which drops to 66.5% with out-of-scope and substantially incomplete responses excluded. The response rates for the previous surveys were higher: 78% (150/192) for 1991, 92% (341/370) for 1994, 85% (505/593) for 1995, 78% (617/787) for 1996, and 76% (450/589) for 1997. Previous surveys were shorter and dealt only with job market issues, whereas the current survey was longer and dealt with many subjects. This may partially explain the lower response rate. Of the 645 usable responses, 602 responses were from practices that are diagnostic radiology only, and 43 responses were from practices that provide both diagnostic radiology and radiation oncology services.
Analysis of potential nonresponse bias was not possible because no information was available on nonrespondents other than practice name and location. In our survey, nonresponse bias might arise at one of two levels: either from the responses to the original mailing used to establish the practice-based sample frame or from responses to the survey itself.
The results from the survey of 1998 hiring were compared with those of the survey of 1997 hiring [21]. The 1997 survey was conducted like the 1998 survey—that is, by mail with mail and telephone follow-up. The principal difference is that the 1998 survey includes some single-radiologist practices, whereas the surveys for 1997 and those for earlier years focused exclusively on multiradiologist groups.
Statistical Analysis
As in the past, each response was weighted. Usable responses were
stratified on the basis of their census region and size of the practice as
measured by total full- and part-time radiologists in the practice at the time
of the survey. The combination of four census regions and seven practice-size
categories resulted in 28 strata for weighting. Strata reflect only those
practices performing some diagnostic radiology services;
radiation-oncology-only groups were excluded throughout the analysis. Each
response was given a weight equal to the total number of practices in the
United States in that stratum divided by the number of responses received from
practices in the stratum. For example, if 30 responses were obtained from a
stratum containing 200 practices, the weight would be calculated as 200
divided by 30, which is equal to 6.67. Weighing responses in this way inflates
the responses to make them match the number of practices providing diagnostic
radiology services, consisting of approximately 2613 multiradiologist groups
and approximately 1666 single-radiologist practices.
Estimates of means, percentages, and totals from this survey, like those from all surveys, are subject to sampling fluctuation. The term "sampling fluctuation" in this context refers to the fact that one may draw many different samples of radiology groups, and for each of them, the results will differ slightly. To measure the sampling fluctuation inherent in the estimates presented here, each estimate is accompanied by its standard error. The standard error is a commonly used measure of the variability associated with the estimate of a particular mean, percentage, or total.
Outcomes compared with those of past years were the total number of positions groups sought to fill, number of individuals retiring or otherwise leaving the profession, and the net excess (or deficit) of positions relative to the estimated number of radiologists seeking positions.
The analysis focuses on multiradiologist groups, with limited information on single-radiologist practices tabulated separately. Relatively few responses were obtained from single-radiologist practices, and uncertainty of their estimated hiring is substantially higher than that for radiology groups.
The total number of positions available was classified into three categories: expansion positions, replacements of radiologists who left the profession (mainly because of retirement or death), and replacements of radiologists who left for another position in the profession (Table 1).
|
Associations between hiring activity and group type, group size, census region, degree of urbanization of the group's location, use of part-time or locum tenens radiologists, and the perceived effect of managed care were also tested (Table 2). The paired association of each group characteristic and hiring activity—whether a group was hiring—was evaluated by the chi-square test.
|
To compare academic groups with private nonacademic groups, we tabulated separately for these two categories of groups some statistics that we expected to differ between the two. In these analyses, nonacademic government groups and miscellaneous other groups were omitted. Both are small categories.
In all the statistical tests used (the Student's t test or the chi-square test), a two-tailed, 95% level of statistical significance was used to classify associations as statistically significant (p < 0.05). With one exception, "significant" is used in this article to indicate statistical significance. The exception is the verbatim reporting of the response "significant" to a survey question about the effect of managed care.
We conducted multivariate regression analyses to ascertain which group characteristics were independently predictive of the total number of positions a group sought to fill and the number of expansion positions a group sought to fill.
For both of these variables, the following group characteristics were examined to determine whether they affected the variable in question: group type (academic, nonacademic government, nonacademic private multispecialty, and nonacademic private radiology); types of services provided (diagnostic radiology or both diagnostic radiology and radiation oncology); use of any part-time radiologists in the past year; geographic region (Northeast, Midwest, South, or West); degree of urbanization of the group's location; group size; and self-assessed effect of managed care on the group.
For the total number of positions to be filled and the total number of expansion positions to be filled, our analysis used the Heckman two-step model [31] to determine, first, if groups that did not have any positions to fill were systematically different from groups that did and to determine, second, which factors affected the number of positions that groups sought to fill if they were hiring.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1998 Findings
In 1998, 51% (±2.1%) of all multiradiologist groups actively
recruited diagnostic radiologists (Table
2), seeking to fill a total of 2299 (±148) positions
(Table 1). Of the 2299
positions in groups, 1167 (±95) or 51% were new positions, 410
(±47) or 18% were positions vacated by radiologists who left the
profession (usually because of retirement or death), and the remaining 722
(±80) or 31% were positions vacated by radiologists who had moved to
another position in the profession.
When all radiology practices are included (group and single-radiologist practices), the estimate for total positions is modestly larger. All practices (group and solo) sought to hire 2647 (±205) diagnostic radiologists. The proportion of positions that were new positions versus replacement positions was similar to that for group practices only.
Multiradiologist groups also reported a total of 422 (±48) vacated positions that they did not seek to fill. Of these, 212 (±40) positions had been vacated by radiologists who left for another position in the profession. This number may include radiologists who were downsized or discharged. The remaining 211 (±34) positions were vacated by radiologists who left the profession (again, principally because of retirement or death). Some of these radiologists may have left a group involuntarily. When all practices are considered (group and solo combined), these figures increase to 590 (±77) departures without replacement, consisting of 281 (±52) individuals who left a practice but remained in radiology and 309 (±47) individuals who left radiology.
Table 2 shows the percentage of multiradiologist groups that actively recruited in 1998, according to group characteristics. (Single-radiologist practices were omitted from this table.) Academic groups were more likely to recruit than nonacademic groups (recall that the academic versus nonacademic distinction was self-designated by respondents), and larger groups were more apt to recruit than smaller groups. Groups located in a nonmetropolitan area were significantly less likely to recruit than others.
Multivariate Analysis of the 1998 Data
Table 2 does not account for
interactions among factors associated with hiring. For example, academic
groups tend to be large and are primarily located in urban areas. Average
hiring for academic groups may reflect the unique aspects of academic
practice, or it may reflect differences in size, location, or other
characteristics between academic groups and other radiology groups.
A more meaningful estimate of the independent effect of size, location, and practice type could be obtained by comparing practices in their peer groups. Academic practices, for example, might be compared with all large, urban nonacademic practices. That contrast—hiring by academic practices versus hiring by an appropriately constructed peer group—would give a better estimate of the independent effect of academic status on hiring.
Multivariate methods use statistical techniques to perform the peer-group comparisons. In the multivariate analysis, hiring for academic groups is compared with hiring predicted for nonacademic groups of similar size and location. The estimate from the multivariate analysis then reflects the independent effect of academic practice, after adjusting for size, location, and other factors entered into the multivariate analysis.
As described in the "Materials and Methods" section, we used the Heckman two-step technique [31] to identify which group characteristics were significant independent predictors of the number of total positions and the number of expansion positions a group sought to fill, with other factors taken into account.
For the total number of positions, the first step of the Heckman technique examined determinants of whether a practice was hiring. In this step, larger groups and groups with both diagnostic radiologists and radiation oncologists were significantly more likely to be recruiting diagnostic radiologists than other groups, if all other characteristics were equal. Academic groups, by contrast, were significantly less likely to be hiring than other groups, if all other characteristics were equal. By geography, the likelihood of hiring was significantly less for groups in the South and West and for groups in the suburbs of large metropolitan areas and those in rural areas, if all other characteristics were equal, than groups in other areas.
The second step of the Heckman technique considered only practices that were hiring and examined determinants of the total number of positions that these groups were seeking to fill. Group size was the only significant determinant of the total number of positions available, all other things equal, with larger groups recruiting more radiologists than smaller groups.
Results from analysis of expansion positions largely paralleled those for total recruitment. Group characteristics significantly associated with higher likelihood of any hiring were also associated with higher likelihood of expansion positions; characteristics associated with lower likelihood of any hiring were also associated with lower likelihood of expansion positions. However, among groups seeking to fill expansion positions, there were no statistically significant determinants of the total number of expansion positions.
Comparison with 1997 and Prior Years
The 1997 survey of hiring practices
[21] included only questions
about hiring in the 12-month period preceding the autumn 1997 date the survey
was completed. The 1998 data shown in this article, by contrast, reflect
hiring for calendar year 1998.
The comparison of 1997 and 1998 hiring data shows that almost all measures of job market activity increased except for departures from the radiology profession (mainly retirements and deaths). In 1997, 44% (±3%) of groups reported recruiting; in 1998, that number rose to 51% (±2.2%). This 1-year increase was just short of statistical significance (p < 0.06), and the 1996 percentage was lower [21]. The total number of openings in multiradiologist groups increased 20%, expansion positions increased 31%, replacements for radiologists who moved to another practice were up 94%, and replacements for individuals who left radiology decreased 37% (Table 3).
|
The total number of openings for diagnostic radiologists in 1998 exceeded the number of such openings estimated in any of the surveys from five previous years [4,6,7,15,21] (Fig. 1 and Table 4). After adjusting the historical data to account for differing time periods (full 12-month year versus partial year-to-date) and target populations (diagnostic radiologists and radiation oncologists combined versus diagnostic radiologists only), estimated total openings for diagnostic radiologists show a U-shaped curve over the decade. From 1991 to 1995, estimated openings for diagnostic radiologists fell by approximately 24%, from 1805 (±192) to 1375 (±117). By 1998, however, total openings had risen to 2299 (±148), which was approximately 27% higher than the 1991 level.
|
|
Comparison of Academic and Private Nonacademic Groups
The total number of positions that academic groups sought to fill was equal
to 15% (±4%) of the total number of radiologists (full-time plus
part-time) in these groups at the time of the survey. For private nonacademic
groups, this ratio was 12% (±1%), which is not statistically
significantly different from academic groups. There was no statistically
significant difference between academic groups and private nonacademic groups
in the ratio of vacated positions the groups did not seek to refill to the
total number of radiologists (full-time and part-time) in these groups at the
time of survey.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With an estimated 2299 positions available in 1998 and approximately 1969 radiologists needing positions, there was a surplus of approximately 330 positions.
This surplus is not statistically significantly different from the estimated 1997 net excess of positions relative to radiologists needing positions of 278. This should be compared with the (not statistically significant) apparent shortfall of 51 positions estimated from the 1996 survey. Thus, the 1998 data show a continuation of 1997's robust job market for diagnostic radiologists.
If positions offered by single-radiologist practices were included in the calculation, the estimated surplus of positions would rise to 465. Although estimated hiring by single-radiologist practices is highly uncertain, this does suggest that excluding solo practices from the analysis gives a conservative (small) estimate of the total surplus of positions available.
Retirements
The number of diagnostic radiologists who left active practice fell sharply
between 1997 and 1998. Results from the 1997 survey indicated that 816
diagnostic radiologists left practice in 1997
[21], and the survey of 1996
hiring showed a similar number
[15]. The 1998 survey, by
contrast, showed 621 (±57) diagnostic radiologists leaving practice,
which is a 24% reduction from 1997 in the number of radiologists leaving
practice. (Specifically, groups reported that in 1998 they sought to replace
410 [±47] diagnostic radiologists who left practice and did not seek to
replace an additional 211 [±34] who ceased practicing; these data added
together indicate that 621 [±57] left practice.) This was a
statistically significant decrease in the estimated number of radiologists
leaving practice.
The number of retirements in 1998 is still substantially higher than the number predicted by nonsurvey methods. Using non-survey-based models [8], we have estimated that approximately 400 diagnostic radiologists would be expected to leave the workforce each year. These models are based on the assumptions that the percentage in each age group (e.g., 50-54 years old, 55-59 years old) who are in active practice (rather than retired) remains constant and that the age-specific mortality rate among diagnostic radiologists equals that of white men in general.
We would expect a reasonable match between the number who left the profession and the predictions of the models because the models are well validated. Specifically, the models have been used to predict with reasonable accuracy the growth over time in the total number of diagnostic radiologists in practice, as reported by the American Medical Association from its "Physician Masterfile" [32]. Also, in a 1995 ACR survey of radiologists [9], no difference was found in the percentage of radiologists in each age category who were retired in 1995 compared with the percentage in 1990.
It appears that a new, large wave of retirements began in about 1996, continued in 1997, and is now tapering off. We suspect that financial and administrative developments in medical practice in the last few years have produced discontent that has led to an increase in the number of retirements. We now have evidence that the high rate of retirement noted in 1997 is not permanent and that retirements have begun to move back toward the level that would be predicted based on the age distribution of active radiologists.
Managed Health Care
Unlike our findings from the surveys of hiring in 1996 and 1997
[15,
21], we found that the
self-perceived impact of managed care had no effect on groups' hiring plans.
In the two other surveys, groups experiencing the greatest self-perceived
effect of managed care were the least likely to have expansion positions or to
offer positions on a partnership-track basis. For 1998, however,
self-perceived effect of managed care had no impact when examined either in
univariate tabulations or in multivariate analysis of hiring.
Both the lack of an impact of managed care and the reduction in retirements suggest that diagnostic radiologists' practices have largely absorbed the changes brought about by managed care. There is no longer a strong association between changes in staffing (hiring and retirement) and the self-perceived impact of managed care on a practice.
Comparison of Academic and Private Groups
The univariate and regression results provide a somewhat different picture
of hiring by academic and nonacademic practices. Univariate results show that,
on average, both academic and private groups had roughly the same number of
openings relative to total group size and that academic groups were more
likely to be hiring. The regression results, by contrast, show that academic
groups were less likely to be hiring, after accounting for their larger size,
urban location, and other factors. The difference between these findings
probably reflects the fact that multivariate analysis takes multiple factors
into account. Academic groups differ from other groups in size and location.
In the univariate results, differences in these factors—for example,
larger size—apparently offset the lower likelihood (for any given size)
of academic groups hiring.
Strengths and Weaknesses of the Data
Our survey data, like all data sets, have strengths and weaknesses. The
strength of our data derives from the large sample (n = 970), which
comprised 23% of all practices performing diagnostic radiology. The high
usable response rate to the survey (66.5%) enhances our confidence in the
findings. A further strength of the 1998 survey procedures is the separation
of radiation oncology from diagnostic radiology. These professions, although
both use radiation, are distinct in their application and in the training of
their practitioners; therefore, separating data on them in an investigation of
hiring is appropriate. Other strengths include a question about vacancies for
which no replacement was sought, the use of a 12-month hiring period, and
follow-up telephone calls to clarify ambiguous responses.
Despite these efforts, some data weaknesses remain. The sampling frame for the survey was based on ACR members' responses to a prior mailing regarding regulatory compliance. Although regulatory compliance affects all practices, we cannot be sure that respondents to our survey form a fully representative cross-section of radiology practices, even after weighting the data to reflect the known distribution of practices by census region and size. For example, practices with no ACR members presumably were missing.
We are also concerned about the declining response rate to the ACR Survey of Hiring. Excessive surveying of groups or a general increase in administrative hassle factor may be making groups less responsive. Although the overall response rate to this survey was high, it was substantially lower than those for prior years. The explanation, however, may be that this survey was much longer than surveys from previous years.
A significant weakness of this survey is the lack of information on groups' success in hiring. Data show the number of openings but do not track the number of interviews and persons hired.
The multivariate analysis of determinants of hiring is subject to substantial uncertainty. Findings from the first stage of the Heckman technique were little affected by minor variations in the set of variables included. Thus, the distinctions between practices that were and were not hiring were reasonably reliably linked to practice characteristics. Results from the second stage of the model, however, were not as consistent. Minor variations in exactly which variables were included sometimes resulted in substantial changes in the estimated impact of the individual practice characteristics used to predict total hiring. This suggests that, among practices that were hiring, there was no strong and systematic linkage between number of positions available in 1998 and a practice's size, location, or type.
Finally, several problems were noted with the information on openings in single-radiologist practices. First, only 56 responses to the survey were for radiologists currently in solo practice. Of these, six reported actively recruiting in 1998. Hence, the estimated level of recruiting by solo practices reflects the responses of only six individuals. Second, because this survey was the first hiring survey for which single-radiologist practices were asked to respond, there is no historical record by which to judge the reliability of the answers to the 1998 survey. Finally, practice size was measured at the time of the survey. Thus, responses regarding openings in single-radioloigst practices must largely reflect unsuccessful attempts to hire. Practices that were once single-radiologist practices that succeeded in hiring would have had two or more radiologists at the time of the survey. For these reasons, data for single-radiologist practices were tabulated separately when used in the analysis.
Acknowledgments
We thank the ACR Radiologist Resources Committee for oversight and helpful
comments.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. S. Lewis, J. H. Sunshine,, and M. Bhargavan A portrait of breast imaging specialists and of the interpretation of mammography in the United States. Am. J. Roentgenol., November 1, 2006; 187(5): W456 - W468. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Merewitz and J. H. Sunshine A Portrait of Pediatric Radiologists in the United States Am. J. Roentgenol., January 1, 2006; 186(1): 12 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine, R. S. Lewis, and M. Bhargavan A Portrait of Interventional Radiologists in the United States Am. J. Roentgenol., November 1, 2005; 185(5): 1103 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bhargavan and J. H. Sunshine Workload of Radiologists in the United States in 2002-2003 and Trends Since 1991-1992 Radiology, September 1, 2005; 236(3): 920 - 931. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. Larson, Y. S. Cypel, H. P. Forman, and J. H. Sunshine A Comprehensive Portrait of Teleradiology in Radiology Practices: Results from the American College of Radiology's 1999 Survey Am. J. Roentgenol., July 1, 2005; 185(1): 24 - 35. [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]
|
|
|
|
|
|
C. Hogan and J. H. Sunshine Financial Ratios in Diagnostic Radiology Practices: Variability and Trends Radiology, March 1, 2004; 230(3): 774 - 782. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Deitch and J. H. Sunshine The Relationship of Managed Care To Business, Professional, and Organizational Aspects of Radiology Practices Am. J. Roentgenol., January 1, 2004; 182(1): 29 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. S. Cypel and J. H. Sunshine Characteristics of Mixed Diagnostic Radiology-Radiation Oncology Practices Am. J. Roentgenol., September 1, 2003; 181(3): 643 - 647. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
J. M. Mitchell and J. H. Sunshine Determinants of Differences Among Radiologists in Starting Salaries Am. J. Roentgenol., May 1, 2002; 178(5): 1067 - 1073. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bhargavan, J. H. Sunshine, and B. Schepps Too Few Radiologists? Am. J. Roentgenol., May 1, 2002; 178(5): 1075 - 1082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine, Y. S. Cypel, and B. Schepps Diagnostic Radiologists in 2000: Basic Characteristics, Practices, and Issues Related to the Radiologist Shortage Am. J. Roentgenol., February 1, 2002; 178(2): 291 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Cooper, T. E. Getzen, H. J. McKee, and P. Laud Economic And Demographic Trends Signal An Impending Physician Shortage Health Aff., January 1, 2002; 21(1): 140 - 154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Sunshine, C. Hogan, and B. Schepps Hiring of Diagnostic Radiologists: An Update Am. J. Roentgenol., December 1, 2001; 177(6): 1484 - 1484. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
