| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Research Department, American College of Radiology, 1891 Preston White Dr.,
Reston, VA 20191.
2 Department of Diagnostic Radiology, Yale University School of Medicine, 333
Cedar St., New Haven, CT 06520.
Received March 13, 2003;
accepted after revision March 28, 2003.
Address correspondence to Y. S. Cypel
(yasminc{at}acr.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. In 1999, the American College of Radiology surveyed 970 practices by mail, using a 65-item questionnaire. A response rate of 66% was achieved. Responses were weighted such that they were representative of all the radiology practices in the United States. The estimates cited in this article were primarily drawn from this 1999 survey.
RESULTS. Mixed practices were nearly evenly divided between large
(area population, 
1 million) and small metropolitan areas (area
population, from 50,000 to 1 million). We found that 63% of mixed practices
were nonacademic private practices; 27% were academic. Approximately 50%
contained 15 or more members. Mixed practices were predominantly owned by
members of the practice and predominantly served both hospitals and
nonhospital settings. At least 90% of mixed practices performed mammography,
imaging-guided breast biopsy, sonography, and nuclear medicine.We found that
compared with multiradiologist diagnostic radiology-only practices, mixed
practices were larger and were more likely to be academic, to serve both
hospital and nonhospital settings, and to perform interventional and nuclear
medicine procedures.
CONCLUSION. The characteristics of a mixed practice differ from those of a multiradiologist diagnostic radiology-only practice.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To address the lack of information on mixed practices, we report in this article the characteristics of mixed practices obtained primarily from data presented in the Survey of Practices conducted in 1999 by the American College of Radiology (ACR) [1]. Data from the study include the number of mixed practices and their regional distribution, location, type and composition, size, and ownership in addition to the types of general practice settings they serve and diagnostic procedures they perform. Our discussion then focuses on how mixed practices differ from the predominant type of radiology practice, the multiradiologist diagnostic radiology-only practice, and the role of mixed practices in today's health care environment.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of the 708 responses received, approximately 90 responses were excluded because they were incomplete, provided by someone other than the intended target group (e.g., residents in training responded but were not considered part of the target population), or duplicated responses already received from a practice. With the irrelevant responses eliminated from the pool of the 970 practices contacted, the final unweighted number of distinct practices from which data for analyses were obtained was 617. These 617 responses constituted a 66% response rate. Forty-one responses (unweighted sample size) were from mixed practices included in the 1999 ACR survey; the unweighted responses were then weighted to produce the estimates presented in this report.
Statistical and Data Analysis Methods
Responses were weighted such that the statistics were representative of the
answers that would have been received if all radiology practices in the United
States had been surveyed and had responded (duplicate responses were each
assigned half the weight otherwise appropriate). We weighted the responses
according to 28 strata on the basis of four census regions and seven
practice-size categories (the latter reflected the total number of full-plus
part-time diagnostic radiologists and radiation oncologists at the time of the
survey). Additional information about how the responses were weighted has been
previously reported [1].
Data from sample surveys are subject to sampling variability, which is a reflection of how much a statistic derived from sample data may deviate from the true underlying population value [2]. The usual measure of sampling variability is the standard error (SE). There is a 95% probability that the true value of a population parameter (such as a population mean) lies within approximately two SEs of the corresponding statistic for a sample drawn from that population. Other descriptive statistics that were computed included frequencies (counts), means (averages), and percentiles (10th, 25th, 50th [also known as the median], 75th, and 90th).
The information presented in the analyses was based on the practice's interpretations of the survey questions and was self-reported. For example, a practice was defined as "academic" if it self-reported its practice type as "primarily academic." Practices also self-reported whether they were diagnostic radiology-only practices, radiation oncology-only practices, or mixed practices (those performing both diagnostic radiology and radiation oncology).
The characteristics of mixed practices were compared with those of multiradiologist diagnostic radiology-only practices, which we defined as those practices having two or more radiologists and performing only diagnostic radiology. We have previously published extensive information about such practices [1], which we labeled as "multiradiologist practices." The term "radiology practices" is used to refer to mixed practices, solo diagnostic radiology-only practices, and multiradiologist diagnostic radiology-only practices.
We used a two-tailed Student's t test to determine whether the difference between two statistics was statistically significant [3]. The estimates used in the statistical comparisons between mixed practices and multiradiologist practices are presented in Tables 1 and 2.
|
|
We used multiple regression analysis to examine the simultaneous effect of multiple independent (explanatory) variables on a continuous dependent variable—a variable that can assume any value along a specified interval for each individual observation. We conducted separate regression analyses to examine which practice characteristics could be used to predict the total number of settings served by a mixed practice and which characteristics could be used to predict the total number of different types of diagnostic procedures performed by a mixed practice. Regression analysis indicates whether a statistically significant association exists between each independent variable and the dependent variable after statistically controlling or adjusting for the effects of all the remaining independent variables. We used a p value of less than or equal to 0.05 (two-tailed) as the measure of statistical significance. All data analyses were conducted using SAS system software (version 8.01, SAS Institute, Cary, NC). If a finding was statistically significant, it was reported as such.
The 1999 ACR survey was weighted according to information gathered in the 2000 ACR Survey of Diagnostic Radiologists and Radiation Oncologists [4, 5]. The most direct, and therefore most accurate, way to estimate the number of practices is to directly use the 2000 ACR survey. We used it for this purpose.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The type of location with the greatest percentage of mixed practices (37%) was the center city of small metropolitan areas (Table 1). Considered more broadly, mixed practices were about evenly divided between large and small metropolitan areas: 44% of mixed practices were located in the center city or suburbs of larger metropolitan areas, and 46% were located in the center city or suburbs of small metropolitan areas.
Other Characteristics of Mixed Practices
Type and composition.—Sixty-three percent of mixed practices
were nonacademic private practices, 27% were academic practices, 8% were part
of nonacademic private multispecialty practices, and 2% were reported as other
types of practices (Table
1).
On average, estimates from the 1999 survey indicated that approximately 83% of radiologists in mixed practices were diagnostic radiologists. The 25th, 50th (median), and 75th percentiles were 79%, 85%, and 90%, respectively.
Size.—The greatest percentage (47%) of mixed practices were those containing 15 or more members (Table 1). Seventeen percent of mixed practices contained 2-4 members, 17% contained 5-10 members, and 20% contained 11-14 members.
Ownership.—Mixed practices were predominantly owned by members of the practice—81% of mixed practices reported being wholly owned by practice members (Table 1). Six percent of mixed practices reported that the practice was partially owned by parties outside the practice, and 13% reported that the practice was completely owned by parties outside the practice. These parties could include ownership by a university or hospital.
Settings served.—Mixed practices predominantly (77%) served both hospitals and nonhospital settings (Table 1). Nineteen percent of mixed practices served only hospitals, and 4% served only nonhospital settings.
A multiple regression analysis with the total number of settings served by a mixed practice as the dependent variable and the characteristics of mixed practices as the independent variables indicated that, after controlling for the effects of all other practice characteristics, practice size was significantly associated (p = 0.03) with the total number of settings served by a mixed practice. On average, every additional five radiologists in a mixed practice increased the total number of settings served by the practice by approximately one. No other characteristics had a statistically significant effect.
Types of diagnostic procedures performed.—The types of diagnostic procedures performed by mixed practices are shown in Table 2. At least 90% of mixed practices performed mammography (98%), sonography (93%), nuclear medicine (91%), and imaging-guided breast biopsy (90%). Other types of procedures commonly performed by mixed practices were CT (89%), interventional (84%), and MRI (82%). For these seven types of diagnostic procedures, the mean total number of procedure types performed by mixed practices was 6.4 (SE ± 0.27). The 10th, 25th, 50th (median), 75th, and 90th percentiles were 5, 7, 7, 7, and 7, respectively.
A multiple regression analysis using the total number of types of procedures performed by a mixed practice as the dependent variable and the characteristics of mixed practices as the independent variables indicated that, after controlling for the effects of all other practice characteristics, being an academic practice was marginally associated (p = 0.05) with the total number of types of procedures performed by a mixed practice. On average, academic mixed practices performed 2.4 more types of procedures than otherwise identical nonacademic private mixed practices.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that there were far fewer mixed practices than multiradiologist diagnostic radiology-only practices. Approximately 2% of all radiology practices (including solo practices) were mixed practices, whereas approximately 62% were multiradiologist diagnostic radiology-only practices.
The percentage of mixed practices that reported being nonacademic private practices was significantly lower than the percentage of multiradiologist diagnostic radiology-only practices: 63% versus 83%, respectively (Table 1) who described themselves as nonacademic private practices.
Approximately 27% of mixed practices were academic practices, significantly greater than the 5% observed for multiradiologist practices. Although the percentage for mixed practices seems large, there are three reasons that the 27% figure is plausible. First, the estimate of approximately 180 academic multiradiologist practices given in the 1999 ACR survey is fairly close to the approximately 204 diagnostic radiologist residency training programs (after osteopathic residency programs are removed) that the education department at ACR [4] knew to exist at that time, which shows that the estimate of the number of academic practices given in the 1999 ACR survey was accurate. Second, with approximately 400 intensively academic hospitals in the United States in the Council on Teaching Hospitals and Health Systems [6], the number of academic mixed practices is small compared with the number of hospitals at which such a program might be based. Third, it is costly for deans of medical schools to establish separate departments for specialties that contain few faculty members. Instead of devoting additional resources for separate departments, administrators have chosen to keep radiation oncologists and diagnostic radiologists in one department (Rogers L, personal communication).
Mixed practices were larger than multiradiologist practices [7]. Practices with 15 or more members comprised 47% of mixed practices compared with 11% for multiradiologist practices (Table 1). Only 17% of mixed practices had 2-4 members, whereas 46% of multiradiologist practices had 2-4 members. As in an earlier study [8], we found that large practices are more likely to offer a broader mix of services than are small practices.
The mean total number of settings served by mixed practices was approximately twice as large (mean, 6.4; SE ± 0.97) as the number served by multiradiologist practices (mean, 3.3; SE ± 0.20). Also, 77% of mixed practices reported serving both hospitals and nonhospital settings, significantly higher than the 56% of multiradiologist practices reported to serve both types of settings (Table 1).
Despite differences in the settings served, both mixed and multiradiologist practices were predominantly member-owned; approximately 85% of each type of practice reported being owned solely by practice members.
The percentage of mixed practices located in the center city of small metropolitan areas was significantly greater than the percentage of multiradiologist practices in this type of location: 37% of mixed practices and 24% of multiradiologist practices (Table 1). Moreover, the percentages of mixed practices situated in nonmetropolitan and rural areas were significantly lower than those found for multiradiologist practices. For nonmetropolitan areas, the percentages were 7% and 19% for mixed practices and multiradiologist practices, respectively. No mixed practices were reported to be in rural areas; yet 12% of multiradiologist practices were rural.
Mixed practices are generally more likely to perform any given type of diagnostic procedure than are multiradiologist practices. We studied seven types of procedures and found statistically significant differences between mixed practices and multiradiologist practices for interventional procedures (84% of mixed practices performing them vs 69% of multiradiologist practices), nuclear medicine procedures (91% vs 80%), and imaging-guided breast biopsy procedures (90% vs 80%) (Table 2). The explanation may be that mixed practices are larger than multiradiologist practices. A pattern that larger practices are more likely to perform various types of diagnostic procedures than smaller practices has been previously reported [1, 7].
The Future of Mixed Practices
Are mixed practices a vanishing vestige of a past when health care was less
specialized and diagnostic radiology and radiation oncology were less distinct
from one another than they now are? Or are mixed practices a small-scale
precursor of the integrated delivery systems that are widely recommended as
the key to better health care in the future?
Much of the evidence points to the former view. For example, in the 1970s, there were approximately 30 physician specialties; by the 1990s, the number had increased to approximately 75 [9]. In addition, from the mid 1930s to the mid 1970s (Cypel YS, Sunshine JH, unpublished data), the American Board of Radiology predominantly issued a general certification for physicians that included diagnostic radiology and radiation oncology; it was reasonable for some physicians to become certified to perform procedures in both fields. Since the mid 1970s, few general certifications have been awarded (Cypel YS, Sunshine JH, unpublished data). Certifications have predominantly been awarded in the distinct specializations of diagnostic radiology or radiation oncology (the latter was renamed from therapeutic radiology to radiation oncology in 1986) (Hammond L, personal communication). Indeed, the process of specialization has continued with the American Board of Radiology, mainly in the 1990s (Hammond L, personal communication), adding multiple certificates of additional qualifications in diagnostic radiology subspecialty fields such as pediatric radiology and vascular and interventional radiology [10].
The Patterns of Care studies begun by the ACR in the mid 1970s found that many of the practitioners of radiation oncology were practicing that specialty part-time and that the quality of patient care (measured by patient outcomes) provided by part-time practitioners was poorer than that provided by full-time practitioners [11]. There was a notable decrease in the number of part-time practitioners of radiation oncology and an increase in the number of full-timers over the ensuing decade [12]. This trend continued subsequently [13], and by the 1990s, physicians practicing both diagnostic radiology and radiation oncology were few [14].
However, these trends are within the scope of activity of individual physicians, not groups, and trends for groups might be different. In fact, the data for groups, which are the most relevant data, clearly show the same trend. A 1991-1992 ACR survey [7] found that approximately 12% of multiradiologist practices were mixed practices, whereas the 2000 ACR survey found that only approximately 3% were mixed practices.
On the other hand, anecdotal evidence [15] based on the reports from one practice suggests that certain mixed practices have important advantages in today's health care environment in which integrated delivery systems are widely recommended. Owners of this large mixed practice contend that their academically based model affords patients a range of particularly high-quality specialized radiologic services, provides greater job satisfaction among practice members because of enhanced ability to work in their areas of specialization, and leads to greater financial stability. Diversification allows slowdowns in certain practice sectors to be compensated for by other sectors. Obviously, large practice size and specialization are important elements to achieving these advantages, and, in this context, we should emphasize that today's mixed practices are relatively large and that a relatively large percentage of them are academic. In conclusion, we believe that most of the small number of mixed practices that currently remain in operation could command many of these advantages and thus be strongly positioned for the future rather than likely to disappear.
Survey Strengths and Limitations
The strengths and limitations of the 1999 ACR survey, the source of most of
the information for this study, have been discussed in recent articles
[1,
16]. A strength that should be
emphasized was the good response rate to the 1999 survey. Despite the
difficulty known to be associated with collecting information from physicians
and other professionals [17]
and the growing workload of radiologists, response rates achieved were
comparable to those for earlier ACR surveys of this type
[18] and other surveys
conducted with physicians
[17]. The use of intensive
telephone follow-up procedures, pretesting, and presentation of the survey
materials and their content contributed to the success of the survey.
Another strength of the survey is that the data included information about mixed practices, a topic that few researchers have specifically addressed to date [7]. This survey and the information on mixed practices obtained from it are therefore important contributions to the existing literature on the characteristics of diagnostic radiology and radiation oncology practices.
The limitations of this study reflect the errors that one may expect with survey research techniques. Because the 1999 ACR survey was based on a sample, it is subject to the errors associated with sample surveys. We attempted to reduce some types of survey errors by conducting extensive pretesting to improve question wording and response options and by targeting issues that were of major importance to the radiology profession [19, 20].
We attempted to reduce nonresponse error (that results when some individuals in the sampling frame do not respond to the questionnaire [21]) by conducting several mailings and then weighting the data received to adjust for nonresponse. Nonetheless, nonresponse is a serious issue. The estimate of the total number of mixed practices reported in the 1999 ACR survey is approximately twice that of the more direct estimate found in the 2000 ACR survey. Apparently, mixed practices were, overall, more likely to respond to the 1999 survey than were other types of practices. If there were differences among practices in the likelihood of responding for which we did not adjust by our weighting by practice size and census region, then the inaccuracies of our data are larger than those indicated by the SEs shown in the tables. SEs always reflect inaccuracies due only to sampling variability, not to nonresponse bias.
The characteristics of mixed practices are typically different from the characteristics found in the predominant form of diagnostic radiology practice, the multiradiologist diagnostic radiology-only practice. Additional research would be useful to improve our understanding of how mixed practices operate in radiology today. Important areas to focus on include the structure of mixed practices, the number of patients that these practices typically serve, the benefits to the practices of establishing ties to research, job satisfaction among practice members, and both professional and patient-based reasons for use of a mixed-practice model over the separate practice of diagnostic radiology and radiation oncology.
Acknowledgments
We thank Barbara Schepps, Chair, ACR Committee on Radiologist Resources,
and J. Frank Wilson, Chair, ACR Commission on Radiation Oncology, for their
thorough reviews, comments, and guidance in the development of this
article.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
|
|
|
|
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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
