HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Saini, S. Articles by Thrall, J. H. Search for Related Content PubMed PubMed Citation Articles by Saini, S. Articles by Thrall, J. H.

Technical Cost of Radiologic Examinations: Analysis across Imaging Modalities¹

¹ From the Department of Radiology, Harvard Medical School, Massachusetts General Hospital, 32 Fruit St, Boston, MA 02114 (S.S., R.T.B., L.A.L., P.F.J., J.H.T.), and the Department of Radiology, Harvard Medical School, Brigham and Women's Hospital, Boston, Mass (S.E.S., P.K., B.F.C.). From the 1997 RSNA scientific assembly. Received June 7, 1999; revision requested July 29; final revision received October 27; accepted November 16. Supported in part by Partners Healthcare System. Address correspondence to S.S. (e-mail: saini.sanjay@mgh.harvard.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To determine the individual technical costs of general diagnostic radiographic, ultrasonographic (US), computed tomographic (CT), magnetic resonance (MR) imaging, and scintigraphic examinations and interventional radiology.

MATERIALS AND METHODS: The Radiology Cost and Productivity Benchmarking Study method of the University HealthSystem Consortium, a cooperative group of academic medical centers, was modified and extended to the six imaging modalities in a tertiary care academic setting. Hospital billing and cost records were analyzed for fiscal year 1996. Costs were divided into labor and nonlabor categories and were allocated to individual imaging modalities on the basis of resources consumed. Physician cost and hospital overhead were not included. Unit costs were analyzed per technical relative value unit (RVU) and per examination.

RESULTS: The costs per technical RVU for diagnostic radiography, US, CT, MR imaging, scintigraphy, and interventional radiology were $65.06, $28.74, $20.95, $17.69, $42.19, and $89.03, respectively. The technical costs per examination for diagnostic radiography, US, CT, MR imaging, scintigraphy, and interventional radiology were $41.92, $50.28, $112.32, $266.96, $196.88, and $692.60, respectively.

CONCLUSION: The method of unit cost analysis for individual imaging modalities was successfully tested in a tertiary care setting. The method should be adopted to allow cost comparison across many institutions, which will permit the promotion of best practices.

Index terms: Cost-effectiveness • Computed tomography (CT), utilization, **.1211² • Economics, medical • Interventional procedures, utilization, **.126 • Magnetic resonance (MR), utilization, **.1214 • Radiography, **.11 • Radionuclide imaging, **.1216 • Ultrasound (US), utilization, **.1298

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Recently, the University HealthSystem Consortium (UHC), which is a cooperative group of academic medical centers, reported on its Radiology Cost and Productivity Benchmarking Study (1). The purpose of that study was to compare radiology departments in university-owned teaching hospitals against radiology departments in university-affiliated teaching hospitals and university-affiliated community hospitals. The investigators in the study, whose findings were published in 1996, analyzed productivity, such as the number of examinations per equipped room, in radiology departments for individual imaging modalities, which comprised computed tomography (CT), ultrasonography (US), magnetic resonance (MR) imaging, scintigraphy, interventional radiology, and general diagnostic radiography. However, costs were analyzed at a gross departmental level. Since this approach provided only a weighted mean for the technical cost of radiologic examinations, the purpose of our study was to extend the UHC cost analysis to individual imaging modalities.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
UHC is an alliance of the clinical enterprises of some academic medical centers and was organized in 1984 to "help members compete in their respective health care markets" (2). In 1995, when the Radiology Cost and Benchmarking Study was undertaken by UHC, there were 101 member hospitals: 49 university-owned hospitals (eg, Stanford University Hospital, Palo Alto, Calif), 34 university-affiliated teaching hospitals (eg, M. D. Anderson Cancer Center, Houston, Texas), and 18 university-affiliated community hospitals.

Our integrated delivery system, which at the time of the UHC study was not a member of the organization, is composed of two university-affiliated tertiary care teaching hospitals. This study was a result of an internal undertaking to compare our institution against similar hospitals. In 1996, our total bed capacity was 1,449, and our total number of outpatient visits was 1,068,032. For the individual imaging modalities, there were 87 general radiography units (including portable radiography machines) with a mean age of 8 years, 20 US scanners with a mean age of 4.5 years, nine CT scanners with a mean age of 2.5 years, three MR imagers with a mean age of 5.5 years, 10 interventional suites with a mean age of 5.8 years, and 21 nuclear medicine scanners (including nuclear cardiology machines) with a mean age of 5.7 years.

In the UHC Radiology Cost and Productivity Benchmarking Study, the spectrum of radiology work was defined on the basis of Current Procedural Terminology, version 4 (CPT-4), codes listed in series 70,000 (all) and 90,000 (selected) (3). CPT-4 codes with an assigned relative value unit (RVU) of zero were excluded from our RVU analysis but were included in the examination analysis. Codes were assigned to one of the six modalities: diagnostic radiography, US, CT, MR imaging, scintigraphy, and interventional radiology. For the majority of codes, this assignment was straightforward. However, in some instances, the assignment was made to allow costs from imaging modalities to be allocated accordingly. For example, biopsy codes for CT and US were assigned to CT and US but not to interventional radiology. We adopted this approach without modification. Therefore, this list of CPT-4 codes served as a basis for identifying the work output for each modality.

For fiscal year 1996, hospital billing office records were queried to obtain a count according to CPT-4 codes of examinations performed from October 1, 1995, to September 30, 1996 (fiscal year 1996). For determining cost, detailed budget office reports were obtained. Since there were differences in billing and budget records between the two hospitals, an independent consulting agency (Integral, Cambridge, Mass) was employed to ensure that the data were complete. Per the UHC method, only the technical component of the Health Care Financing Administration RVU scale was considered.

The UHC method involved the division of costs into labor and nonlabor categories, as shown in the Figure. We too adopted this approach, and all costs incurred by the radiology department were assigned to a specific modality. For this, statements from the hospital's budget office were used. In cases in which resources were not shared across modalities, the assignment of production costs to individual modalities was straightforward. For example, the costs of technologists working in CT were allocated to CT costs. However, when resources were shared, such as nursing, the assignment of cost was made on the basis of resources consumed, as shown in Table 1. For equipment costs, either the annual lease expense or the purchase price amortized over a period recommended by the American Hospital Association, Chicago, Ill, was used.

View larger version (22K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Lists of labor and nonlabor cost components and cost-allocation methods. Film = image, IS = information systems, QA = quality assurance, RN = nursing, % = percentage of.

An important difference between our study and the UHC study is that the UHC survey required 12 months for analysis, which led to variation in the reporting period among the contributing institutions. In addition, whenever complete data were not available, hospitals were allowed to estimate the total work output on the basis of the "highest-volume" examinations, as long as they represented at least 65% of the total procedure volume. In our study, the overall work output and cost figures were exact, although their allocation to individual modalities did have some imprecision because when common resources were used for more than one modality, an estimate was made for the actual usage for a given modality (eg, nursing).

Per the UHC method, costs were analyzed on a per RVU basis. The resource-based relative value scale was developed to pay for physician services by indexing medical procedures to an arbritrary RVU scale (4). The cost-per-RVU approach allows cost comparison across imaging modalities or institutions; the differences in cost may be due to differences in procedures. From the RVU scale, the malpractice component and practice component were excluded, which allowed the determination of cost per technical RVU. However, because the RVU scale is inexact, we also determined technical costs on a per examination basis.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Total work output and unit costs for each modality are shown in Table 1. The mean RVU per examination was 2.13, and the RVU per examination ranged from 0.64 for diagnostic radiography to 15.09 for MR imaging. The mean cost per technical RVU was $37.86, and the cost per technical RVU ranged from $17.69 for MR imaging to $89.03 for interventional procedures. The cost per technical RVU for diagnostic radiographic studies was more than three times that for MR imaging and CT. On a per examination basis, costs were lowest for diagnostic radiography and US and were highest for interventional procedures.

The breakdown in costs per examination for different inputs is provided in Table 2. Overall comparison of mean costs per technical RVU for our study against those for UHC institutions is provided in Table 3. Note that the UHC data were for 1995 while the data from the present study were for fiscal year 1996. Furthermore, in the UHC study, transcription and patient transportation costs were excluded because of inconsistencies in the data—hence the slightly lower total cost per technical RVU in Table 3 in comparison with Table 1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Our purpose in presenting our findings is not only to report our results but also to describe our method so that future comparative analyses can be performed to identify the best practices. Although the cost-accounting method developed for this analysis was applied in a tertiary care hospital, the approach should be transferable to other types of institutions since cost drivers are similar. Indeed, the UHC report included university hospitals and community hospitals.

Our results provide some interesting insights on work output and costs in radiology departments. While approximately 55% of the departmental costs were in general diagnostic or interventional radiology, these studies contributed only about 30% of a department's technical RVUs. In contrast, CT and MR imaging accounted for more than 52% of the department's technical RVU output while contributing only about 29% to the department's costs. For US and scintigraphy, the relationship was appropriate, with similar percent costs and percent technical RVUs.

Since reimbursements are related to RVUs, these findings have important implications on "profitability." Specifically, because hospital-based departments are obliged to perform the full spectrum of radiologic services, CT and MR imaging services appear to subsidize procedures in general and interventional radiology since the overall cost per RVU approximates the Health Care Financing Administration conversion factor.

The analysis also suggests that the RVU assignment may be incorrect, which may result in improper conclusions when costs are compared on a per RVU basis. For example, in our study, MR imaging had a high mean technical RVU per examination of 15.09 compared with that of 7.78 for interventional radiology. As a result, the cost per technical RVU for the two modalities was reversed, with a low cost basis ($17.69) for MR imaging and a nearly fivefold higher cost basis ($89.03) for interventional radiology.

Indeed, it was this observation that led us to compare costs on a per examination basis. Our results suggest that the per examination method may be more useful because a technical cost per examination for each of the various modalities is easier to understand, and our findings do make intuitive sense. For example, the total cost per interventional radiologic examination was approximately 2.6 times higher than the total cost per MR imaging examination with this method ($692.60 vs $266.96 [Table 1]). However, we did not formally study the accuracy of the RVU scale, and it is possible that other factors may explain the findings.

Thus, when a benchmark is used to compare costs across institutions for individual modalities, the cost per technical RVU and the cost per examination may be necessary. When costs are compared across modalities, the cost per examination may be preferable. This, for example, will allow one to determine how much more expensive a CT examination is compared with an intravenous urographic examination for the evaluation of flank pain (present study: $112.32 vs $41.92).

Regardless, one limitation of either approach is that, just as a cost on a departmental level is too coarse, with the tendency to average costs for different modalities, analysis on a modality level is also limited since it too requires averaging of the costs of protocols that have different degrees of complexity. For example, the cost of an MR imaging examination without contrast material will differ markedly from the cost of a study with contrast material because of the high cost of MR imaging contrast media and the additional postcontrast imaging time. Hence, a more useful analysis would provide the cost of a nonenhanced study separately from the cost of a contrast medium–enhanced study. Thus, future studies will need to provide more detailed cost analysis within each modality for the data to be useful in controlling costs or in determining the cost-effectiveness of competing radiologic procedures (5).

Analysis of cost components within each modality is useful for defining opportunities for cost control and productivity improvement. For example, as expected, technologist wages, equipment and its servicing, contrast media, and film are the important cost drivers for all modalities, with medical-surgical supplies and nursing also contributing importantly in the case of interventional radiology.

While the obvious implication is that costs can be reduced by focusing on each of these areas, it is equally important to see where additional expenditures can be made to make the work flow more efficient. An apparent opportunity appears to be in the area of patient transportation. As can be seen in Table 2, on a per examination basis, patient transportation costs are very low: less than $1 for CT and MR imaging. Even if these costs are allocated exclusively to inpatient examinations (which makes transportation costs about $3 per examination because inpatients represented about one-third of the total patients who underwent CT or MR imaging examinations), the expenditure remains low. Since delays in patient transportation services frequently lead to decreases in productivity, our results suggest that hospitals may be underfunding these internal services.

Indeed, a doubling of resources will not substantially add to examination costs, and improved productivity will lead to lower overall unit costs. Improved productivity reduces costs when fixed costs can be spread out over more examinations.

However, it is important to note that cost minimization is not the sole objective of medical services. Obviously, quality parameters must also be considered. For example, since imaging equipment is underused on weekends and at night, one approach to increase productivity and reduce costs is to offer imaging services during these periods of low utilization. While many outpatients welcome the possibility of weekend services, inpatient services performed during the overnight shift degrade the overall patient experience. Thus, technical costs alone should not be a determining factor of procedural selection.

This article provides insight only on cost, and we emphasize that our purpose was not to undertake a cost-effectiveness analysis. Indeed, the implications of our findings will vary depending on the situation. For example, utilization decisions might differ under capitation and fee-for-service scenarios. More interesting, regardless of the technical cost considerations, providers may make utilization decisions on the basis of their own work flow considerations. For example, clinicians may prefer abdominal CT over US because the CT scans are easier than the US scans for them to review. Finally, in addition to medical effectiveness, the value-based approach requires that many quality parameters such as patient experience be also considered. We anticipate that these more comprehensive analyses will be the focus of future studies.

An important limitation of our results is that the cost of physician services and hospital overhead costs have not been included. Each of these cost components requires its own comprehensive analysis similar to the one performed here for technical costs. This was not performed in this analysis and was therefore excluded from the cost formula. However, these are important expenditures that will contribute to the overall cost of radiologic procedures. However, in spite of this limitation, our results are relevant because if the choice is between two radiologic studies, physician costs and hospital overhead charges will not be eliminated and may well be quite similar.

Finally, it should be noted that the idea of benchmark comparison is not to compare individual performance against mean performance of peers, as is done in Table 3, but to compare individual performance against best-in-class performance. Since these data are not widely available, such comparisons are difficult. The UHC report does identify academic institutions with high productivity (and therefore presumably lower unit costs). For example, when technical RVUs per equipped room are examined for MR imaging, our mean productivity of 93,755 greatly exceeds the UHC mean of 42,087 but pales in comparison with the best-in-class performer, the productivity of which was 162,296 (1). Thus, an important objective of the radiology community should be to work collaboratively, perhaps through organizations such as the American College of Radiology, Reston, Va, to undertake an ongoing data collection process so that best practices can be identified. If the operational processes of best-in-class performers can be adopted widely, radiologic procedures will become more cost-effective and highly competitive in the overall health care market.

	FOOTNOTES

 
**. Multiple body systems

Abbreviations: CPT-4 = Current Procedural Terminology, version 4, RVU = relative value unit, UHC = University HealthSystem Consortium

Author contributions: Guarantors of integrity of entire study, S.S., S.E.S., R.T.B., J.H.T.; study concepts and design, S.S., S.E.S., R.T.B., J.H.T.; definition of intellectual content, S.S., S.E.S., R.T.B., J.H.T.; literature research, S.S.; data acquisition, P.F.J., L.A.L., B.F.C., P.K.; data analysis, S.S., L.A.L., P.K.; manuscript preparation and editing, S.S.; manuscript review, S.E.S., R.T.B., J.H.T., L.A.L., P.F.J., P.K., B.F.C.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Radiology Cost and Productivity Benchmarking Study Oak Brook, Ill: University HealthSystem Consortium, 1996.
2. University HealthSystem Consortium Web site; Available at: http://www.uhc.org. Accessed July 1999..
3. Kirschner CG, Davis SJ, Jackson JA, et al. Physicians current procedural terminology: CPT '97 Chicago, Ill: American Medical Association, 1996.
4. Hsiao WC, Braun P, Dunn D, Becker ER. Resource-based relative values: an overview. JAMA 1988; 260:2347-2353.[Abstract/Free Full Text]
5. Hillman BJ. New imaging technology and cost containment. AJR Am J Roentgenol 1994; 162:503-506.[Abstract/Free Full Text]

This article has been cited by other articles:

				A. Singh, R. Danrad, P. F. Hahn, M. A. Blake, P. R. Mueller, and R. A. Novelline MR Imaging of the Acute Abdomen and Pelvis: Acute Appendicitis and Beyond RadioGraphics, September 1, 2007; 27(5): 1419 - 1431. [Abstract] [Full Text] [PDF]

				W. Hollingworth Radiology Cost and Outcomes Studies: Standard Practice and Emerging Methods Am. J. Roentgenol., October 1, 2005; 185(4): 833 - 839. [Full Text] [PDF]

				D. T. Gray, W. Hollingworth, C. C. Blackmore, M. A. Alotis, B. I. Martin, S. D. Sullivan, R. A. Deyo, and J. G. Jarvik Conventional Radiography, Rapid MR Imaging, and Conventional MR Imaging for Low Back Pain: Activity-based Costs and Reimbursement Radiology, June 1, 2003; 227(3): 669 - 680. [Abstract] [Full Text] [PDF]

				W. W. Mayo-Smith, J. T. Rhea, W. J. Smith, C. M. Cobb, I. F. Gareen, and G. S. Dorfman Transportable versus Fixed Platform CT Scanners: Comparison of Costs Radiology, January 1, 2003; 226(1): 63 - 68. [Abstract] [Full Text] [PDF]

				M. E. Liebeskind, P. H. Arger, A. Liebeskind, K. Maston, and C. Langlotz Using Sonography to Examine Adult Patients at an Academic Medical Center: Have Usage Patterns Changed with the Expansion of Managed Care? Am. J. Roentgenol., December 1, 2002; 179(6): 1395 - 1399. [Abstract] [Full Text] [PDF]

				M. D. Cohen Determining Costs of Imaging Services Radiology, September 1, 2001; 220(3): 563 - 565. [Full Text] [PDF]

				H. P. Forman Cost, Value, and Price: What Is the Difference and Why Care? Radiology, January 1, 2001; 218(1): 25 - 26. [Full Text]

				W. W. Mayo-Smith Administration of a CT Division Radiology, February 1, 2002; 222(2): 319 - 326. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Saini, S. Articles by Thrall, J. H. Search for Related Content PubMed PubMed Citation Articles by Saini, S. Articles by Thrall, J. H.