| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Thomas Jefferson University, 1096 Main Bldg,, 132 S.
10th St., Philadelphia, PA 19107.
2 Research Department, American College of Radiology, 1891 Preston White Dr.,
Reston, VA 20191-4397.
Received October 26, 2001;
accepted after revision January 22, 2002.
Address correspondence to D. G. Mitchell.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. CT and MR imaging procedure volumes for pelvis, abdomen, chest, and total were extracted from the 1993, 1996, and 1999 Health Care Financing Administration Physician/Supplier Procedure Summary Master Files, based on CPT-4 codes. We also analyzed a fee-for-service health insurance database for January 1998 through July 1999 from a single northeastern state, which included provider location (rural, suburban, or urban) and type (teaching or nonteaching site).
RESULTS. The greatest 3-year Medicare increase was for obdominal MR imaging, from 1996 to 1999 (101% increase). However, pelvic, abdominal, and chest MR imaging together remained less than 5% of total MR imaging. Abdominal MR imaging increased more than did total MR imaging in all 10 Health Care Financing Administration regions. In the fee-for-service database, the relative procedure volume of abdominal MR imaging varied approximately fivefold from rural to urban provider locations, and approximately double from nonteaching to teaching hospitals.
CONCLUSION. Although far more abdominal CT than abdominal MR imaging is performed, the rate of abdominal MR imaging utilization has increased more rapidly since 1993. The relative procedure volume of abdominal MR imaging varied more than fivefold from rural to urban provider locations and double from nonteaching to teaching hospitals.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this article, we explore data from two separate insurance databases with regard to various categories of body MR imaging and CT and for total MR imaging and CT volume. At the start of our study, we hypothesized that the variations in MR imaging equipment and training of radiologists would manifest as regional variation in the utilization of body MR imaging. We also hypothesized that recent technical innovations have led the utilization of body MR imaging to grow faster than that of other MR imaging applications or that of body CT. However, we suspect that the technical capabilities and expertise for routine use of body MR imaging have not diffused extensively from academic centers into the community; therefore, we hypothesized that body MR imaging utilization would vary depending on the type of provider.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We were interested in the pelvic, abdominal, chest, neurologic (brain, head or neck, and spine) and extremity MR imaging and CT CPT-4 codes. These codes include MR imaging abdomen: 74181, 74185; MR imaging pelvis: 72196, 72198; MR imaging chest/cardiac/breast: 71550, 71555, 75552, 75553, 75554, 75555, 75556, 76093, 76094; MR imaging other: 72141,72142, 72146, 72147, 72148, 72149, 72156, 72157, 72158, 72159, 73220, 73221, 73720, 73721, 76400, 73225, 73725, 70336, 70540, 75041, 70551, 70552, 70553; CT abdomen: 74150, 74160, 74170; CT pelvis: 72192, 72193, 72194; CT chest/cardiac: 71250, 71260, 71270; CT other: 70450, 70460, 70470, 70480, 70481, 70482, 70486, 70487, 70488, 70490, 70491, 70492, 73200, 73201, 73202, 73700, 73701, 73702, 72125, 72126, 72127, 72128, 72129, 72130, 72131, 72132, 72133.
For each group of procedures, we calculated procedure volume, relative percentage (number of procedures in the category of interest relative to all MR imaging or CT performed), and the utilization rate per 100,000 beneficiaries. Chest and breast were considered together because many breast MR imaging examinations may have been coded as examinations of the chest according to reimbursement policies. Codes for CT follow-up, CT guidance, image reconstruction, and MR spectroscopy were not counted because the body part examined could not be determined.
In addition to a count of the whole database, we also examined the volume on a regional basis and calculated percentages and rates. The 10 HCFA regions, named after the city in which the regional office is located (Atlanta, Boston, Chicago, Dallas, Denver, Kansas City, New York, Philadelphia, Seattle, and San Francisco), were used for regional analysis. Changes in relative procedure volume between 1993, 1996, and 1999 were also examined.
The summary Medicare database did not provide information regarding specific providers. Therefore, we conducted a separate analysis on a fee-for-service health insurance database from a single northeastern state, encompassing 19 consecutive months (January 1998-July 1999). The records included provider name, location, and radiology group affiliation. The database covered an average of 227,307 patients per month, 95% of whom were under the age of 65. The demographics of the patient population are summarized in Table 1. MR imaging and CT procedure volumes were calculated for the categories described. This data set provided the location and name of the provider of service. Using provider zip code, we categorized providers as rural, suburban, or urban. The three largest providers, each of which maintained radiology residency training programs, were categorized as academic teaching hospitals and were compared with the other 32 providers.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
A far greater number of pelvic, abdominal, and chest studies were performed using CT than were performed with MR imaging; the difference was well over an order of magnitude. For example, 1,895,746 abdominal CT examinations were performed in 1993, compared with only 26,763 abdominal MR imaging procedures. Even after abdominal MR imaging increased at a rate more than three times that of abdominal CT (114.7% vs 28.0%), the volume of abdominal CT remained more than 40 times that of abdominal MR imaging (2,425,911 vs 57,450).
From 1993 to 1996 and from 1996 to 1999, the volume of neurologic CT increased at approximately half the rate of total CT. Extremity CT increased by 45.1% from 1993 to 1996, but then decreased by 27.2% during the next 3 years and remained a small portion of total CT. The volume of neurologic MR imaging increased at rates slightly less than those of total MR imaging, but the increase in extremity MR imaging was approximately double that of total MR imaging.
Among body CT procedures, CT of the pelvis and chest increased most, with an increase of slightly more than 60% between 1993 and 1999, which was approximately double the increase of total CT. Abdominal CT increased by 28% during this period, slightly less than the overall increase in total CT. In distinction, abdominal MR imaging showed a much greater increase than did the other body MR imaging procedures. The increase in abdominal MR imaging was almost entirely restricted to the interval between 1996 and 1999, during which time it increased by more than 100%, in contrast to its minimal increase of 6.6% from 1993 to 1996. Pelvic MR imaging increased by approximately one third from 1993 to 1999, approximately half the rate of increase of total MR imaging. In 1993 and 1996, MR imaging was performed more commonly for the pelvis than for the abdomen. By 1999, abdominal MR imaging had become more common. MR imaging of the chest (including breast and cardiac studies) decreased by 13.6% from 1993 to 1996, but it increased by 10.1% from 1996 to 1999.
The procedure volume for imaging the pelvis, abdomen, chest, neurologic structures, and extremities, relative to total CT and MR imaging procedure volume, and the changes in these rates are presented in Table 3. In 1993, CT of the pelvis, abdomen, and chest together accounted for more than half of all CT, whereas imaging of these body parts accounted for only 5% of total MR imaging. During the entire 6-year period from 1993 to 1999, the greatest increase in relative rate was for extremity MR imaging, 45.1% (from 11% to 16% of total MR imaging), resulting from increases of approximately 20% in each of the two 3-year intervals. The greatest single 3-year increase was in the relative rate for abdominal MR imaging, which increased 49.7% from 1996 to 1999 (an increase from 1.5% to 2.2%), compared with an increase for abdominal CT of only 1.6% for that period. Among abdominal studies performed using these two modalities, only 1.4% was performed using MR imaging in 1993 and 1996, increasing to 2.3% in 1999. The smaller change in the proportion of examinations using CT indicates that abdominal CT increased at a pace corresponding to the increase of CT in general. Modest increases occurred in the relative rates of pelvic and chest CT—13.4% and 14.1%, respectively. However, relative rates for pelvic and chest MR imaging decreased 3.9% and 18.1%, respectively: although the actual volume of these two procedures increased, the increase of total MR imaging procedures was greater. The relative rate of neurologic MR imaging decreased slightly, by 1.6% from 1993 to 1996 and then by 4.1% from 1996 to 1999.
|
For all Medicare regions, the volume of all categories of CT and MR imaging increased, but at different rates. The regional percentage of total MR imaging procedures for abdomen for 1993, 1996, and 1999 is presented in Figure 1. The utilization of abdominal MR imaging relative to all MR imaging decreased for nine of 10 regions between 1993 and 1996, but it increased substantially for all 10 regions between 1996 and 1999, ranging between 22.3% and 119.6%. During this period, the relative percentage of pelvic MR imaging changed less, varying from -18.8% to 21.4%, and chest, cardiac, and breast MR imaging decreased in all but one region.
|
State Insurance Claims Database
The total volume of non-body procedures performed was similar for CT and MR
imaging—20,512 versus 25,720, respectively. However, a marked
discrepancy was noted in the number of pelvic, abdominal, and chest
examinations performed using CT and MR imaging
(Table 4). The percentage
relative to total modality for pelvic, abdominal, and chest procedures was
approximately 10 times higher for CT (20.9%, 27.0%, and 10.9%, respectively)
than for MR imaging (2.0%, 1.8%, and 0.7%, respectively).
|
The relative percentages of CT studies of the pelvis, abdomen, and chest to total CT studies were similar for rural, suburban, and urban locations (Fig. 2). For example, the relative percentage for abdominal CT was 24.4%, 25.6%, and 28.1%, respectively, at rural, suburban, and urban locations. However, MR imaging rates showed marked variation, especially for MR imaging of the abdomen (Fig. 3). In rural locations, only five abdominal MR imaging procedures were performed, accounting for 0.3% of the total MR imaging procedures. On the other hand, abdominal MR imaging was 1.0% of the total in suburban and 3.1% in urban locations. A similar but smaller trend was noted for pelvic and chest procedures.
|
|
For teaching versus nonteaching hospitals, only slight differences were found in the relative percentage of CT studies of the pelvis, abdomen, and chest (Fig. 4). Considered together, CT of the pelvis, abdomen, and chest accounted for 60.3% and 58.4% of total CT procedures performed at teaching versus nonteaching hospitals, respectively. However, the proportion of MR imaging studies directed to the pelvis, abdomen, and chest at teaching institutions (7.0%) was double that at nonteaching institutions (3.5%) (Fig. 5).
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several factors may influence the much lower proportion of MR imaging than CT for imaging of the pelvis, abdomen, and chest. CT was already an established method for imaging these body parts before MR imaging was introduced as a clinical examination. Although MR imaging has shown considerable usefulness, practice patterns have not changed greatly, and most physicians still consider CT to be the preferred method for most body cross-sectional imaging. Additionally, the diffusion of equipment, software, and expertise involved in conducting and interpreting CT examinations is far more advanced than for MR imaging.
Compared with the Medicare database, the single-state fee-for-service database included a smaller number of patients in one particular region rather than the entire country, but it allowed us to compare MR imaging and CT usage depending on the type of provider. This analysis showed that body MR imaging frequency, especially for the abdomen, is higher in urban centers and at teaching hospitals. Although patient populations at urban centers and teaching hospitals presumably differ from those at nonurban and nonteaching centers, the relative percentage of body CT procedures between the two types of centers differed far less than it did for MR imaging procedures. Most likely, body MR imaging is used more commonly at centers where equipment and the level of subspecialty training are more favorable.
We suspect that most abdominal and pelvic MR imaging is performed in or near university centers or other centers of expertise, and that diffusion into the community is at an early stage. Another factor that may affect the current preference of CT over MR imaging for examining the abdomen and pelvis may be a perception that CT is more cost-effective. We are not aware of an analysis for either of these sites that actually shows this to be the case, taking into account the cost of procedures that might be required after the performance of each of these examinations before a final diagnosis and patient disposition is accomplished.
Our study analyzed two databases. The selection of patients and the methods of tabulating data differ greatly between the two. The patient population is especially different between the Medicare population, who are mostly older than 65 years, and the single-state fee-for-service population, who are nearly all younger than 65 years. Utilization rates and volumes derived from these two databases may not be directly comparable to each other. It is also uncertain as to how accurately the billing records of a single large insurance provider in one state reflect the actual number of procedures performed in that state or how closely these figures resemble national trends. However, in spite of these differences between the two databases, we note that relative use of body CT and body MR imaging appeared similar.
In conclusion, national Medicare data show that although far more abdominal CT than abdominal MR imaging is performed, the rate of abdominal MR imaging utilization has been increasing more rapidly than that of abdominal CT since 1993. A single-state provider database shows that the relative procedure volume of abdominal MR imaging varied greatly: more than fivefold from rural to urban provider locations and approximately double from non-teaching to teaching hospitals. Variation was somewhat less for pelvic and chest MR imaging, and it was minimal for abdominal, pelvic, and chest CT. Greater diffusion of up-to-date equipment and training in body MR imaging are likely to lead to its increased use when indicated. This conclusion is consistent with national and regional Medicare data showing that abdominal MR imaging has consistently increased in relative usage, and the increase may indicate the beginning of this diffusion.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. P. Krestin, J. C. Miller, S. J. Golding, G. G. Frija, G. M. Glazer, H. G. Ringertz, and J. H. Thrall Reinventing Radiology in a Digital and Molecular Age: Summary of Proceedings of the Sixth Biannual Symposium of the International Society for Strategic Studies in Radiology (IS3R), August 25 27, 2005 Radiology, September 1, 2007; 244(3): 633 - 638. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
