Hostname: page-component-7b9c58cd5d-9klzr Total loading time: 0 Render date: 2025-03-15T17:33:32.093Z Has data issue: false hasContentIssue false
  • English
  • Français

Marginal cost of operating a positron emission tomography center in a regulatory environment

Published online by Cambridge University Press:  25 October 2005

Anderson Chuck ,
Philip Jacobs ,
J. Wayne Logus ,
Donald St. Hilaire ,
Chester Chmielowiec  and
Alexander J. B. McEwan
Anderson Chuck
Affiliation:
University of Alberta
Philip Jacobs
Affiliation:
University of Alberta and Institute of Health Economics
J. Wayne Logus
Affiliation:
Cross Cancer Institute
Donald St. Hilaire
Affiliation:
Cross Cancer Institute
Chester Chmielowiec
Affiliation:
University of Alberta and Cross Cancer Institute
Alexander J. B. McEwan
Affiliation:
University of Alberta and Cross Cancer Institute
Rights & Permissions [Opens in a new window]

Abstract

Objectives: Cost studies of positron emission tomography (PET) imaging are important for resource and operational planning; the most relevant cost analysis in this regard is the marginal cost. Operating within a regulatory environment can add considerably to the costs of providing PET services. Previously published research has not examined the marginal cost structure of PET nor have they described the implications of regulatory compliance to operational costs. The purpose of this study was to conduct a comprehensive cost estimation of PET imaging with 18F-fluorodeoxyglucose (18F-FDG) to better identify the fixed and variable cost components, the marginal cost structure, and the added costs of satisfying regulatory requirements.

Methods: Financial data on capital and operating expenses were collected for the PET center at the Cross Cancer Institute in Edmonton, Alberta, Canada.

Results: The total per-service cost for clinical operations ranged between $7,869 (400 annual scans) and $1,231 (3,200 annual scans). The marginal cost for the center remained steady as volume increased up to the throughput capacity.

Conclusions: Results indicate that economies from increased volumes did not arise. Regulatory requirements added significant costs to operating an 18F-FDG-PET center.

Keywords

Positron emission tomographyNuclear imagingCosts
Type
GENERAL ESSAYS
Information
International Journal of Technology Assessment in Health Care , Volume 21 , Issue 4 , October 2005 , pp. 442 - 451
Copyright
© 2005 Cambridge University Press

Positron emission tomography (PET) is an advanced diagnostic imaging technique (11) used for the detection and management of a wide variety of conditions, including cancers (10;13), heart disease (3), and neurological disorders (2). Despite its wide clinical diagnostic applicability (1;14;15), operating a PET center remains extremely complex and is widely perceived to be expensive (2;4).

Although there have been many studies that explore the clinical applications of PET (3;7;14), most research investigating the costs of PET has been less comprehensive, with studies differing in design and economic assumptions. Considerable variation in the configuration of the PET centers (6), the goals of these centers and the regulatory environment governing PET services further complicate the analysis. These variables may have a significant impact on associated costs.

Because the majority of PET costs are fixed (5), throughput volume has a major impact on the annual cost per scanning procedure. Fixed costs will be spread out more with higher volumes (1). However, the fact that unit fixed costs are declining with volume does not provide guidance with regard to resource planning whose focus is on how many additional resources are added as volume increases. The relevant concept in this regard is the marginal cost, which is the additional cost required to produce one more unit of output (i.e., additional cost to conduct additional scanning procedures) (8). Some cost studies of PET have reported fixed and variable costs (2;5;9) but even these have not specified in any great detail these cost components nor have they examined the marginal cost of providing PET services.

In the case of PET, marginal costs inform the allocative decisions concerning the consequence of adding additional scanning procedures within the range of a center's throughput capacity. Thus the concepts of fixed and marginal costs are central to planning and management decisions: fixed costs must be used if one is to plan for a new unit. Once that unit is in place, the planner must know how much extra it will cost for additional usage. With technologies such as PET, new applications and clinical indications are continually being developed for existing facilities, and it is the added resource requirements that are important when evaluating these new applications.

An additional component of supplying PET services in the current regulatory environment is the added cost of complying with regulatory standards for this technology. PET operates in a highly regulated environment that will vary between countries and even regions within countries. In the Canadian healthcare environment, this issue is particularly acute as the recent imposition of a new regulatory framework (http://www.hc-sc.gc.ca/hpfb-dgpsa/tpd-dpt/ctd_cta_notice_e.html) has imposed a considerable increase in the cost of operating a PET facility, and considerable complexity to operational issues.

According to the Eur-Assess methodology subgroup report (12), health technology assessments should take into account the regulatory environment in which the technology is going to operate. Thus the purpose of this study was to conduct a comprehensive cost estimation of 18F-FDG-PET to more clearly explicate fixed and variable cost components, the marginal cost structure, and the added costs of regulatory compliance.

Facility and Setting

A typical PET center that produces its own positron emitting isotopes requires a facility capable of housing the production and scanning infrastructure, a cyclotron to synthesize the positron-emitting isotopes, automated synthesis units to produce the radiopharmaceutical, a PET scanner(s), and image review and data archiving hardware and software (2). In 2001, the Cross Cancer Institute (CCI) in Edmonton, Alberta, Canada, constructed an entirely new facility to house a cyclotron and clinical chemistry laboratory to supply PET radiopharmaceuticals to two scanners housed in the Division of Nuclear Medicine. The CCI is the major facility in Northern Alberta operated by the Alberta Cancer Board, which provides cancer care to a population of approximately 1.8 million. This facility currently provides 18F-FDG and clinical imaging services for the Cross Cancer Institute and through this facility to patients resident in Alberta. It is anticipated that it will shortly commence supply of 18F-FDG to the neighboring city of Calgary and to other locations in Edmonton and Western Canada in the future.

Using this new laboratory facility to produce clinical FDG—and ultimately other research-based PET radiotracers—allows routine clinical and research PET imaging to be performed on the two scanners. Current patient throughput is limited by budgetary constraints, but this issue will be resolved in future budget years.

It is possible to define three main phases associated with the conduct of an 18F-FDG-PET scan. The first phase consists of fluorine-18 (18F) production followed by synthesis of injectable 18F-FDG (FDG Synthesis). The second phase consists of conducting quality control on the 18F-FDG, which involves compliance with current Good Manufacturing Practice Standards (cGMP) (FDG Preparation), and the third consists of conducting the PET imaging procedure. Once the scan is completed, there is also the final responsibility of image interpretation and issuing a report (Clinical Imaging).

Regulatory Environment

A critical cost component is the added costs of complying with regulatory standards. Regulatory agencies around the world increasingly have been moving toward a more stringent regulatory environment for radiopharmaceuticals, despite decades of evidence of safe and effective use. In this context, as a late arrival in the field of FDG production, our group believed it important to establish a facility that was compliant with the framework of good manufacturing practice regulations (cGMP) for the production of radiopharmaceuticals in general, and for FDG in particular. In Canada, the regulations governing clinical trials and cGMP have been interpreted to mean that all seven sites in the country that produce FDG are regarded as manufacturers of the product, with cGMP requirements comparable to those expected of large pharmaceutical manufacturers. Manufacturing files have been prepared, and appropriate regulatory submissions have been initiated (http://www.hc-sc.gc.ca/hpfb-dgpsa/tpd-dpt/ctd_cta_notice_e.html). Some of the facility construction costs reflect these requirements. This regulatory environment will ultimately mean seven new drug submissions across the country for seven identical products.

The regulatory environment has also required that each site is regarded as a clinical trial sponsor, with requirements to establish good clinical practice (cGCP) infrastructure. The societal cost implications of this regulatory approach are significant, and the costs at a local level are now very much part of the cost of establishing a PET program. We, therefore, have included them in this analysis.

MATERIALS AND METHODS

Study Design

Costs associated with the 18F-FDG-PET center at the CCI since 2001 were collected from the financial records of the Alberta Cancer Board. Within each of the three main components associated with the conduct of a 18F-FDG-PET scan, the costing framework was composed of four steps: (i) identifying resources; (ii) categorizing resources into fixed and variable components; (iii) measuring resource use within the range of throughput volume; and (iv) costing measured resources.

Infrastructure Expenditures

The PET center at the CCI was established as a new program, with new construction providing cyclotron and cGMP laboratories and renovations allowing siting of the two scanners. The costs of construction and renovations were identified as actual costs from the construction budget data. The cost of new equipment required for operating the facility was identified from purchase orders. The items associated with construction, and with equipment purchases are itemized in Table 1. Capital costs for equipment and facilities were amortized over appropriate periods of time for each piece of equipment and discounted at a rate of 10 percent over their lifetime to obtain an annual cost for each capital item. The scanners were amortized over 7 years and the cyclotron over 20 years.

Operational Costs

Fixed and variable costs were analyzed and adjusted for expected annual utilization by clinical and research operations. Accordingly, operational costs were obtained by identifying personnel and resources that were required to operate the cyclotron and produce 18F-fluoride, synthesize 18F-FDG, conduct appropriate quality control activities of cGMP, and maintain regulations of cGCP. In addition, the clinical resources that were needed to scan the patients and interpret the images were obtained by observing the actual resources and costs of operating the scanners. Costs were based on 2004 levels.

Annual Throughput

Annual throughput capacity was determined by multiplying the number of available scanning days per operating year by the number of scanning procedures per day. Imaging hours for the center were typically 9:00 AM to 5:00 PM Tuesday to Friday, with Mondays reserved for preventative maintenance of the cyclotron and PET scanners. Moreover, there are 52 weeks in a year, and excluding holidays and planned and unplanned downtime, there were approximately 50 possible procedure weeks per year. Thus, with 4 procedure days per week, the operational time available for imaging was estimated at 200 scanning days per operating year.

Typical daily throughput volume for clinical and research operations was estimated by analyzing patient volumes. Typically, clinical operations conduct twelve scanning procedures per operating day (shared between the two scanners). It was estimated that ten research procedures per week were also performed. The typical annual throughput volume was, therefore, 2,400 scanning procedures for clinical operations and 500 scanning procedures for research use.

Maximum daily throughput volume for clinical operations was estimated by analyzing the time required to conduct a whole body scan (providing a more conservative estimate of yearly throughput) on each PET scanner. With a 10-minute interval between scans, there can be a maximum of one scanning procedure every 50 minutes on each of the two scanners. Therefore, with approximately eight imaging hours per operating day associated with a single shift, there is a maximum of sixteen scanning procedures per operating day, giving an estimated maximum throughput of 3,200 scanning procedures per year.

Per-Service and Marginal Cost Analysis

The total cost for providing 18F-FDG and clinical imaging services was derived by adding together the total cost (variable and fixed) of each phase. To obtain an annual cost per service, the total cost was divided by the center's yearly typical and maximum throughput capacity.

To determine the marginal cost, total annual costs for providing 18F-FDG and clinical imaging services at subsequently higher levels of throughput were compared, resulting in the change in cost associated with changes in volume. That is, the marginal cost at a given throughput level is calculated by dividing the difference in cost between the current and previous throughput level with the difference in throughput levels [marginal cost=(total cost at given output–total cost at previous output) ÷ (given output–previous output)]. For decisions affecting shorter time periods, this amounts to focusing on variable costs, because fixed costs will remain unchanged for smaller changes in volume.

Cost Assumptions

Resources for research and clinical operations are analogous as they use the same equipment and are distinguished by requirements for extra scanning time and extra nurse, physician, and technologist involvement. For this analysis, it was assumed that all research studies were performed using 18F-FDG. Additionally, although the number of required 18F-FDG doses for any given operating day is variable, there is an assumption in this manuscript that only one 18F-FDG production run will be conducted per operating day; this run is completed by 9:00 AM, with production of the activity required to meet the scheduled number of daily doses, with appropriate decay correction.

The cost driver for FDG Synthesis was the timing of the scheduled scanning procedures because greater time periods between synthesis and scan required synthesizing greater amounts of FDG to account for isotope decay, resulting in an increased consumption of resources. The variable costs components, therefore, were the cyclotron operator, supplies, and utilities (power). For estimating the cost of the imaging component, the cost driver selected was the number of annual scanning procedures. Variable cost components were the imaging nurse, imaging technologists, nuclear medicine physician, and supplies. There was no specific cost driver identified for 18F-FDG Preparation, given that quality control is conducted on each production batch of 18F-FDG and resource consumption remains fixed.

RESULTS

Costs

Identified resources and their sources of valuation are presented in Table 1. Table 2 shows the annual cost for capital expenditures. All costs are in Canadian dollars. The total fixed operating cost was $2,911,896 (Table 3). The total variable cost was $783,169 at the typical throughput volume (Table 3) and $1,026,484 at the maximum throughput volume. Conducting 500 annual research-related procedures added an incremental cost of $132,937. Regulatory compliance added an additional $625,925 to total costs (approximately 20 percent), all of which were fixed costs. Costs associated with cGMP and cGCP were $352,822 and $273,103, respectively.

Within the range of possible throughput volumes (400 and 3,200 annual procedures), total costs ranged between $3,147,747 and $3,938,380 (Table 4). The marginal costs remained steady up to 2,800 annual scanning procedures at approximately $276 per patient scanned and significantly increased to $332 at 3,200 annual scanning procedures (Table 4). This finding means that an increase in volume from 2,400 to 2,500 scans would result in increased total costs of $27,000 ($277×100).

Per-Service Cost and Sensitivity Analysis

It is important to mention that research operations added an additional 500 scanning procedures per operating year to the existing level of clinical throughput. At the typical annual throughput volume, the total cost per scanning procedure was $1,540 for clinical operations and $1,320 when adding research operations (Table 4). At the maximum annual throughput volume, the total cost per scanning procedure was $1,231 for clinical operations and $1,100 when adding research operations.

The cost driver selected for FDG Synthesis was the timing of the scheduled scanning procedures. To account for the variability of resources consumed due to the timing of scanning procedures, a one-way sensitivity analysis was conducted to determine the estimated cost per scan for 1000 annual scanning procedures (5 scans per day) varied within the range of possible successively scheduled procedures for a given operating day. It takes 180 minutes to complete five scanning procedures between the two clinical scanners. The earliest that five scanning procedures can be scheduled is 9:00 AM to 11:00 AM, giving an estimated cost per scan of $2,837. The latest that five scanning procedure can be scheduled is 1:30 PM to 3:30 PM, giving an estimated cost per scan of $2,863.

DISCUSSION

Cost studies of PET are important for planning services and operations (9) and the most relevant cost analysis in this regard is the marginal cost. This analysis provides a comprehensive cost estimation of not only developing and operating a PET center with on-site radiopharmaceutical manufacturing—and research expectations—but better identifies both fixed and variable cost components, the marginal cost behavior, and the added costs of complying with regulatory standards.

Construction and Equipment Costs

The cost of constructing and equipping a new facility is included in this analysis as this component is critical to the planning process and has not been well developed in previous analyses. This cost will be dependent upon the goals of the center. If the intent is to produce only FDG for local use, the total space requirements would be less than developed for the facility at the CCI. However, many of the cost drivers of this type of facility are not linearly related to facility size; the costs of constructing a facility compliant with radiation safety and cGMP requirements are relatively independent of size, and so we believe that our calculations are accurate and broadly comparable to the construction costs associated with a generic PET facility.

The annualized amortized construction costs for the CCI facility are $298,943. These are costs that have been difficult to derive from previous papers; we believe they are robust enough to be included in future planning processes for the construction of PET imaging and production facilities. These annualized costs reflect the complexity of planning and building a facility housing an operational cyclotron.

Equipment Costs

The scanners purchased for our operations include one PET and one PET-computed tomography (CT) scanner; costs associated with two PET-CT scanners would obviously be greater but so would anticipated throughput. The remainder of the equipment purchases were related to the costs associated with FDG Synthesis, FDG Preparation, and Clinical Imaging. The total annualized and amortized costs associated with equipping our facility were $1,646,330. These are significant but not considerably greater than the cost associated with equipping any stand-alone imaging facility and certainly with the broad costs associated with hospital and radiology department equipment costs.

Operational Costs

The operational costs associated with personnel, consumables, and regulatory compliance was $1,225,288. These costs included a full-time physicist to support the program as we believed it important that a physicist's contributions be recognized as critical to the successful, and safe, operation of a PET center. These costs reflect the assumption of a single daily synthesis and do not reflect costs associated with failed bombardments or failed FDG syntheses. The complexities of operating two PET scanners at high throughput are reflected in the staffing complement proposed in Table 3 and are broadly reflective of staffing complements discussed in the literature.

The total cost, therefore, of operating entirely as a clinical center at typical throughput volumes was $3,695,065. Conducting 500 research-related procedures added $132,937. Of this figure, the additional annual cost of complying with regulatory requirements was $625,925 (includes cGMP=$352,822 & cGCP=$273,103). We believe these figures to be robust as initial budgetary calculations for this facility were made before the imposition of the new Canadian regulations; if our calculated additional regulatory costs are excluded from this analysis, the figures closely mirror the initial calculations. These costs have been defined as only those required to comply with the requirements of the new regulations and not with costs associated with routine quality control and quality assurance.

As expected, the total (fixed and variable) cost per scanning procedure decreased with higher throughput volumes for all levels of operation, which had a range between $7,869 (two scans per day) and $1,231 (sixteen scans per day) for clinical operations. The decrease in total cost per scanning procedure at higher levels of throughput volume is a direct result of the high fixed costs that, in the present analysis, comprise between 74 percent and 92 percent of the total costs for clinical operations. At increased levels of production, marginal costs remain steady and only increase when approaching operational capacity. That marginal costs only increase near full capacity and do not decrease at higher levels of throughput volume suggests that economies (i.e., increases in efficiency) from increased volumes will not arise. However, near full capacity significant diseconomies will occur, because marginal costs begin to increase. As additional capacity is added—more synthesis and imaging procedures—marginal costs will again remain relatively fixed until full capacity is again reached. It is important to note that marginal costs reflect a relatively small proportion of total costs but do have an important bearing upon the planning process.

Comparing these results to previously published research is difficult due to the considerable variation in the configuration of PET centers. Still, in one of the more comprehensive cost analysis of PET that also reported costs on quality control (9), the estimated cost per scan of providing PET services with on-site radiopharmaceutical manufacturing (including a dedicated cyclotron and a single scanner) for whole body scanning procedures was $1,659 USD at 1,488 annual scans. In the present analysis, for between 1,200 and 1,600 annual whole body scanning procedures, the estimated cost per scan ranged between $2,171 and $2,803 CAD (includes dedicated cyclotron and two scanners). At more realistic annual throughput volumes for a facility with two scanners (3,200 scans), the cost per scan falls to $1,231, and at the volumes claimed by industry of at least 2,000 annual scans, the cost per scan would be $1,792. However, it is probable that these volumes are unrealistic in routine clinical practice with a mix of patient acuities. Noting that this cost comparison represents data collected from two different health care systems, the identified cost components between our studies are comparable. However, the total cost of complying with regulatory standards of cGMP was significantly greater in the present study ($273,103 CAD compared $80,000 USD), and reflects the considerable international variations in regulatory requirements for the production of positron emitting radiopharmaceuticals. However, it should be noted that the authors did not report the costs associated with oversight and governance functions (cGCP), which are critical elements of the operating environment in most countries. The Canadian regulations may be considered to be a reflection of the increasing regulatory expectations of PET facilities and may be a harbinger of a world-wide increase in these expectations and costs.

The regulatory environment is adding significant costs to operating a PET center, and this finding is exemplified by the experience of the Canadian community. The current regulatory environment in Canada treats each PET center with a cyclotron as a manufacturing site, requiring what are essentially university-based research groups to develop the same regulatory infrastructure as large radiopharmaceutical companies for producing what is essentially identical products. Although all PET centers with on-site radiopharmaceutical manufacturing require appropriate safeguards to ensure the quality and safety of their radiopharmaceuticals, PET centers within Canada have the added expectations of new regulations; these are similar to those imposed by a recent European Union directive. In the present analysis, regulatory compliance added $196 to the cost of a scan (assuming 3,200 scans per year); cGCP costs alone added a total of $273,103 (or $95 per scan) to the annual operating cost of the PET center. However, these costs only reflect the cost of personnel. The true cost of cGCP and cGMP is greater when including the added costs of office space, equipment, data storage, and supplies. Sustaining this level of cost in this environment is doubtful with effects already being observed in delays in the clinical introduction of the technology, lost research opportunities and productivity, and a loss of international competitiveness.

These results need to be examined in light of study limitations. This analysis did not include costs for data archiving, because this cost element is difficult to separate out from other archiving procedures. Electronic and hard copies of each scanned image need to be archived and stored appropriately, which not only add additional facilities and equipment but significant variable costs in terms of supplies and personnel. Another limitation was that the cost of utilities for FDG Synthesis was likely underestimated because actual cost data were not available and conservative estimated costs were used instead. Furthermore, while no model can perfectly represent actual conditions (9), the economic assumptions adopted for this analysis were designed to be both accurate and straightforward for easier interpretation of results. A sensitivity analysis was used to address areas of uncertainty and in particular the relationship between costs and the timing of scheduled scanning procedures. The sensitivity analysis indicated that the cost per scanning procedure was not sensitive to the timing of scheduled scanning procedures.

CONCLUSION

Our study is a comprehensive cost estimation of operating a PET facility, including radiopharmaceutical supply and imaging, and to represent those costs that require consideration before establishing a PET imaging and/or cyclotron facility. Our results have practical implications for PET managers and directors, because it more clearly demonstrates how variable costs change within an 18F-FDG-PET center's operational capacity. Specifically, it allows for a more valid examination of the resulting rise in cost for each additional scanning procedure (unit of output). This type of analysis becomes even more critical in light of the increasing acceptance of PET as a diagnostic technology and expanding use as a research tool and will be particularly relevant as new PET radiopharmaceuticals enter widespread clinical use. The potential configuration and complexity of PET centers are expanding to include more sophisticated equipment (9). Planning of services for these complex PET centers is essential, given the constant increase of political and budgetary constraints in the health-care system. PET center managers and directors will increasingly need to provide credible evidence to justify business plans as they are developed.

It is, however, instructive to note that, at throughputs of 3,200 patients per year, the costs associated with operating a PET facility with two scanners remains economical and within the realm of other high technology medical procedures. What these data do provide is reassurance that, when appropriately planned and operationalized, PET imaging can be performed economically and that it is possible to provide funders and planners with accurate and reproducible costs for establishing these core facilities.

There are societal implications associated with the regulatory costs associated with the introduction of clinical PET services, although the possible repercussions on the PET research community may be more severe. Admittedly, our study is in a Canadian context, but with harmonization being the key goal of regulatory agencies worldwide, these issues are likely to take on widespread applicability. It is important to appropriately assess risk when establishing a regulatory regimen. Meaningful discussion between the providers of PET services and government is needed to facilitate an appropriate balance between safety, efficiency, and research.

CONTACT INFORMATION

Anderson Chuck, MPH (), Doctoral Student, Department of Public Health Sciences, University of Alberta, 13-103 Clinical Sciences Building, Edmonton, Alberta T6G 2G3, Canada

Philip Jacobs, DPhil (), Professor, Department of Public Health Sciences, University of Alberta, 13-103 Clinical Sciences Building, Edmonton, Alberta T6G 2G3, Canada; Institute of Health Economics, 1200-10405 Jasper Avenue, Edmonton, Alberta T5J 3N4, Canada

J. Wayne Logus, MSc (), Research Associate, Donald St. Hilaire, RTNM, RTMR (), Professor, Manager, Department of Oncologic Imaging, Cross Cancer Institute, 11560 University Avenue, Edmonton, Alberta T6G 1Z2, Canada

Chester Chmielowiec, MD (), Assistant Clinical Professor, Department of Radiology & Diagnostic Imaging, University of Alberta, 8440 112 Street, Edmonton, Alberta T6G 2B7; Department of Oncologic Imaging, Cross Cancer Institute, 11560 University Avenue, Edmonton, Alberta T6G 1Z2, Canada

Alexander J. B. McEwan, MB (), Department of Oncology, University of Alberta; Department of Oncologic Imaging, Cross Cancer Institute, 11560 University Avenue, Edmonton, Alberta T6G 1Z2, Canada; Department of Oncologic Imaging, Cross Cancer Institute, 11560 University Avenue, Edmonton, Alberta T6G 1Z2, Canada

References

Berger M, Gould MK, Barnett PG. 2003 The cost of positron emission tomography in six United States veterans affairs hospitals and two academic health centers. AJR Am J Roentgenol. 181: 359– 365.Google Scholar
Conti PS, Keppler JS, Halls JM. 1994 Positron emission tomography: A financial and operational analysis. Am J Roentgenol. 162: 1279– 1286.Google Scholar
Eitzman D, Al-Aouar Z, Kanter HL, et al. 1992 Clinical outcome of patients with advanced coronary artery disease after viable studies with positron emission tomography. J Am Coll Cardiol. 20: 559– 565.Google Scholar
Evens RG, Seigel BA, Welch MJ. 1983 Ter-pogassian MM. Cost analyses of positron emission tomography for clinical use. AJR Am J Roentgenol. 141: 1073– 1076.Google Scholar
Evens RG, Seigel BA, Welch MJ, Ter-pogassian M. 1983 Cost analyses of positron emission tomography for clinical use. AJR Am J Roentgenol. 141: 1073– 076.Google Scholar
Frick MP, Gupta NC, Sunderland JJ, et al. 1992 Considerations in setting up a positron emission tomography center. Semin Nucl Med. 22: 182– 188.Google Scholar
Gardner SF, Green JA, Bednarczyk EM, et al. 1992 Principles and clinical applications of positron emission tomography. Am J Hosp Pharm. 49: 1499– 1506.Google Scholar
Jacobs P, Rapaport J. 2002 The economics of health and medical care. 5th ed. Gaithersburg, Maryland: Aspen Publishers, Inc.;
Keppler JS, Conti PS. 2001 A cost analysis of positron emission tomography. AJR Am J Roentgenol. 177: 31– 40.Google Scholar
Kwee SA, Coel MN, Lim J, Ko JP. 2005 Prostate cancer localization with 18fluorine fluorocholine positron emission tomography. J Urol. 173: 252– 255.Google Scholar
Lassen U. 2001. Positron emission tomography (PET) with 18-F-fluorodeoxyglucose (FDG): A survey of the literature with regard to evidence for clinical use in oncology, cardiology and neurology. Alberta, Canada: Centre for Evaluation and Medical Technology Assessment;
Liberati A, Sheldon T, Banta D. 1997 Eur-assess project subgroup report on methodology: Methodological guidance for the conduct of health technology assessment. Int J Technol Assess Health Care. 13: 186– 219.Google Scholar
Ryu SY, Kim MH, Choi SC, et al. 2003 Detection of early recurrence with 18F-FDG PET in patients with cervical cancer. J Nucl Med. 44: 347– 352.Google Scholar
Strauss LG, Conti PS. 1991 The application of PET in clinical oncology. J Nucl Med. 32: 623– 648.Google Scholar
Wagner HN, Conti PS. 1991 Advances in medical imaging for cancer diagnosis and treatment. Cancer. 67: 1121– 1128.Google Scholar
Figure 0

Source of Data and Valuation for Identified Resources

Figure 1

Estimated Annual Costa of Capital Expenditures Relative to Their Estimated Life Time

Figure 2

Annual Fixed and Variable Costs for CCI 18F-FDG-PET Center at Its Typical Throughput Volume for Clinical Operations

Figure 3

Marginal Costs, Estimated Per-Service Cost, and Sensitivity Analysis