Costs of plain-film radiography in a partially digitized radiology department

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Costs of plain-film radiography in a partially digitized radiology department J. Alanen, L. Keski-Nisula, J. Laurila, I. Suramo, C.-G. Standertskjöld- Nordenstam & M. Brommels To cite this article: J. Alanen, L. Keski-Nisula, J. Laurila, I. Suramo, C.-G. Standertskjöld- Nordenstam & M. Brommels (199) Costs of plain-film radiography in a partially digitized radiology department, Acta Radiologica, 39:, To link to this article: Published online: 04 Jan 010. Submit your article to this journal Article views: 46 Full Terms & Conditions of access and use can be found at

2 Acta Radiologica 39 (199) Printed in Denmark - All rights reserved Covvriaht 6 Acta Radioloaica 199 ACTA RAD I0 LO GI CA ISSN COSTS OF PLAIN-FILM RADIOGRAPHY IN A PARTIALLY DIGITIZED RADIOLOGY DEPARTMENT An activity-based cost analysis J. ALANEN~, L. KESKI-NISULA, J. LAURILA~, I. SURAMO, C.-G. STANDERTSKJ~LD-NORDENSTA~ and M. BRoMMELs Department of Clinical Radiology, Oulu University Hospital, Oulu; Department of Radiology, Vaasa Central Hospital, Vaasa; 3Department of Radiology, Porvoo District Hospital, Porvoo; 4Department of Radiology, Helsinki University Central Hospital, Helsinki; and 5Department of Public Health, University of Helsinki, Helsinki; Finland. Abstract Purpose: The aim of the study was to analyse the costs of computed radiography (CR) as part of a small picture archiving and communication system (mini-pacs), and to compare these costs with those of conventional analogue radiography using activity-based accounting (ABC). Material and Methods: The study was conducted at the Central Hospital of Vaasa where in 1993 the Radiology Department acquired a mini-pacs with a CR reader, a chest CR unit, and a CT unit as digital image processing modalities. Of altogether plain-film examinations, 3/4 were made with CR and stored mostly on film, and 1/4 were made with conventional analogue radiography. The costs and activities of these two modes were analysed by means of the ABC method which identifies and allocates indirect costs in radiological procedures. Results: The costs of CR imaging were 9% higher than those of conventional radiography. The costs of the chest CR unit were equal to those of conventional radiography. The difference was due to higher investment costs in digital image processing. The time gained from a reduction in the number of retakes did not shorten the time spent by patients in the examination room, and its effect on film costs was minimal. Conclusion: In planning the step-by-step transition of conventional filmbased analogue radiography to fully digitized radiography, it should be noted that films are still used in the transition period and that this is associated with higher costs than in the previous system of conventional analogue plain-film imaging. Key words: Radiography, picture archiving and communication systems; costs and cost analysis. Correspondence: Leo Keski-Nisula, Department of Radiology, Vaasa Central Hospital, Hietalahdenkatu 4, FIN Vaasa, Finland. FAX Accepted for publication August The last decade has seen the rapid increase in the use of the digital radiographic technique and many radiology departments have installed a picture archiving and communication system (PACS). These systems vary from large, hospital-wide installations aimed at the filmless hospital to mini-pacs which link a few workstations to CT, MR, and a local digital image archive. Parallel to this development, it has also become common to use computed radiography (CR) without PACS in skeletal and chest radiography. Owing to economic restrictions on health care in Europe, the transition from a conventional film-based system to a digital filmless hospital in one step is usually impossible. Hence 00

3 COSTS OF PLAIN-FILM RADIOGRAPHY Table 1 Radiological examinations: figures for one year Plain-film radiography Ultrasonography CT Fluoroscopy Mammography Angiography Interventions Total n YO (approx.) the change must be performed step by step. The proportion of digital modalities has gradually increased over the years and still seems to be increasing, despite a higher initial investment compared to conventional equipment. Arguments used in favour of acquiring digital equipment point to the benefits such as lower costs, rationalization of technical operations, and higher quality in the produced images. The effect that digital plain-film radiography has on the operating costs of a radiology department has been examined mainly in hospitals where radiological imaging is almost totally digitized (1,, 5, 7, 1, 14). Cost analyses have also been published on the use of CR for intensive care unit patients (6,, 19). However, reports on the cost of using CR in skeletal and chest radiography are few and their results contradictory (1,, 5-7, 1-14, 19). As a rule, these analyses have been based on traditional cost accounting. Since the spring of 1993, the Radiology Department of the Central Hospital of Vaasa has used an imaging system in which skeletal and chest radiographs are stored on phosphorus plates, and the information contained in them can thus be used on workstations. In the majority of cases, the reporting is done on film. Conventional film-screen radiography is also used, parallel to this system. The purpose of our study was to assess the costs of CR and to compare these with the costs of conventional film-screen radiography. Material and Methods The Central Hospital of Vaasa is a medium-sized general hospital with approximately 500 beds. In 1994, the Radiology Department performed about examinations (Table 1); (75%) of them were skeletal and chest radiographies of which approximately 3/4 were carried out with CR, and 1/4 with conventional film-screen radiography (Table ). The digital imaging modalities of the Radiology Department are connected to one image archive and two laser printers (3M) by an image distribution net. There are two workstations: one with programs designed for processing and interpreting CRs (Siemens Magic View 110); the other designed specifically for working with CT. A third workstation is located in the Intensive Care Unit (ICU), and functions as an image-viewing workstation (Siemens Litebox). The digital image information is transferred to this third workstation from the CR reader and chest CR-unit. At the workstation, the radiographer can postprocess the images if necessary and print them on 34x43 cm laser film. Several images of a patient are combined in one am. The radiologist usually reports on the examination from the film, and uses the workstation if a more detailed analysis is necessary. Images of the ICU patients are sent from the Radiology Department to the workstation in the ICU. The images are stored on the net at three levels: on the workstations; on the archive server; and on optic disks in a jukebox. Images are searched for in the system files if the information printed on film is not sufficient or if the film is not available for interpretation. There is also a film archive for hard-copies. Table Plain-film radiography: figures for one year n % (approx.) Examinations at the Radiology Department CR (other than chest CR examinations) Chest CR unit examinations Conventional plain-film radiography 4570 Radiography outside the Radiology Department Bedside CR Conventional plain-film radiography at the Oncology Department Unclassified I Total

4 J. ALANEN ET AL. Activity accounting model Fig. 1. Principles of ABC accounting. The resources are allocated to resource pools. The examination is divided into activities to which resources from the resource pools are allocated according to cost drivers. Activity costs are allocated to products (examinations) using activity drivers. One or several resource pools can be utilized for an activity, and an activity can be part of one or several examinations. The radiographs from conventional film-screen examinations are developed in a daylight processor (Agfa Curix Capacity). Cost analysis This study is concerned with the costs of all the plain radiographic examinations (plain-film examinations) made in the Radiology Department during the year The examinations were divided into three groups: CR (other than chest CR examinations); chest CR examinations; and conven- tional plain-film radiography. The distribution of the examinations was equal in the groups of computed and conventional plain-& radiography. The costs were analysed by means of activity-based cost accounting (ABC) (4). The principles of ABC are summarised in Fig. 1. All costs directly and indirectly related to activities in the Radiology Department were accounted for. Costs were calculated in Finnish marks (FIM; 1 USD=approximately 5.10 FIM). The costs of the resources (personnel, equipment, materials, premises, general costs, etc.) were collected into resource pools each of which contained functionally or spatially related resources. For example, one pool was called chest imaging room and comprised the costs of the premises, the chest imaging equipment, and the materials used in the examination room. The resources were allocated to pools according to real use. Each separate activity necessary for carrying out an examination was defined. Costs were allocated to activities according to resource use as expressed by cost drivers (Table 3). Cost drivers for personnel, equipment and premises were defined as the time required for the activity, multiplied by the frequency of this activity; and cost drivers for materials as the actual amount used. The number of personnel required was taken into account when personnel costs were allocated. The information on average time and number of personnel required was collected in a 5-month follow-up study using our internal radiology depart- Table 3 Allocation from resource pools to activities, using plain-film CR as an example Activity Resource pool Cost driver Time reservation and registration, waiting Main procedure (examination) Image processing Activity with the CR reader Image processing at the workstation Hard-copy printing Image interpretation and reporting Film handling and transport in the Radiology Department Clinical film demonstration Quality control Other personnel (secretary) Examination room Radiographers Radiography equipment Radiographers Equipment Material Radiologist Other personnel Equipment Other personnel Viewing boxes Radiologist Viewing boxes Radiographers Equipment 5min 5min 10 min 10 min 10 min min as used 10 min min 9min 9 min 9min 5min lmin 1 min 0

5 COSTS OF PLAIN-FILM RADIOGRAPHY Table 4 Costs of plain-jilm radiography Capital Personnel Film and Other Total chemicals FIM %;I* FIM %>* FIM %* FIM %I* FIM Yo* CR (other than chest CR examination) Chest CR unit examination Conventional plain-film radiography * Percentages are approximate ment information system. Other information concerning cost and consumption was obtained from the hospital Accounting Department and from inventory files. The activity-based costs were allocated to the examinations in proportion to the numbers performed. The Radiology Department budget for 1994, including capital costs, was FIM 19.4 million. Of this sum: personnel costs accounted for FIM 7. million; material costs for FIM 5.5 million; maintenance costs and rent for FIM 1. million; and the capital and maintenance costs of equipment for FIM 4. million. The total value of the equipment in the Radiology Department was FIM. million. The maintenance and updating costs of programs used in the digital equipment were included in the annual operating costs. Acquisition costs were distributed over the period The mini-pacs, the CT scanner, and plain-film radiography equipment were all purchased in 1993 at a cost of FIM 9.0 million. The capital investment depreciation was calculated in equal installments. The amortization period of all radiography machines was 15 years, except for the roentgen tubes for which it was 3 years. Three cost comparisons were performed. In the baseline calculations, the amortization period of the mini-pacs was set at 10 years and the interest rate for invested capital was defined as OYO. In addition, we carried out a sensitivity analysis in which the amortization period for the mini-pacs was set at 5 and 15 years, and the interest rates at 4% and % respectively. Results The baseline cost analysis (defining the mini-pacs amortization period at 10 years and the interest 0 CI] 0 El Quality control 0 Clinical film demonstration 0 Film handling and transport Image interpretation and reporting Image processing 0 Ln 0 Main procedure (examination) Fig.. Costs of activities in plain-film radiography..time reservatioi and registration 0 L CR Chest CR Conventional unit plain-film 03

6 J. ALANEN ET AL. Time reservation and registration Main procedure (examination) Image processing Image interpretation and reporting Film handling and transport Clinical film demonstration Quality control Total * Percentages are approximate. Table 5 Costs of activities in plain-jzm radiography CR Chest CR unit Conventional plain-film (other than chest CR) examination radiography FIM YO* FIM %* FIM %* CR (other than chest CR) Chest CR unit examination Conventional plain-film radiography Table 6 Costs of image processing Capital Personnel Film and Other Total FIM FIM chemicals FIM FIM FIM rate on invested capital at 0%) showed the CR examinations to have total costs that were 9% higher than those of conventional plain-film radiography (Table 4). The costs of the chest CR unit were equal to the costs of conventional radiography. The capital costs of digital imaging were almost twice as high as the capital costs of conventional imaging. Personnel costs were lowest in chest CR imaging, but approximately the same in the other digital imaging modes as in conventional imaging. Film costs were equal in all groups despite the higher price of laser film used in digital imaging. The costs of premises, allocated hospital costs, and administrative costs were similar in both digital and conventional imaging, although slightly lower in chest CR. The higher cost of digital investigations were due to the activity termed image processing, in which the phosphorus plates were loaded into the reader unit, processed on the workstation, and printed out on film by a laser printer (Fig., Table 5). The cost of this activity was FIM 39 per examination (including FIM 1 in capital costs), i.e. 56% higher than the processing of a conventional radiograph for which the cost was calculated at FIM 5 per examination (including FIM 3 in capital costs) (Table 6). The use of the workstation and the net as well as the archive amounted to about FIM 1 per examination. On average, 1.5 films per skeletal or chest examination could be saved by using the workstation to combine several images on one film sheet. Thus FIM 6 per examination could be saved by working with a workstation, compared to working without a workstation. This represents approximately 34% of the total cost of each examination. The costs of a radiographer s work and the running of an examination room (including equipment) were allocated to the activity termed main procedure. The time the patient spent in the examination room and the number of examinations completed were used as cost drivers. On average, the patient occupied the room for 10 min for CR and conventional imaging, and 7 min for chest CR. Despite these time differences, the costs of the main procedure (examination) for conventional and digital plain radiography were similar (Table 5). The effect of the length of the amortization period of capital goods on the cost of the examination was analysed by defining the amortization period for the mini-pacs equipment and the daylight processor used in conventional imaging as 5, 10 and 15 years; and by defining the interest rate as 0% (Fig. 3). A shorter amortization period increased the cost of digital imaging more than the 04

7 COSTS OF PLAIN-FILM RADIOGRAPHY ation. An interest rate of 4% resulted in a 1% (FIM 19) difference, and an interest rate of % in a 15% (FIM 5) difference between digital and conventional radiography. i 5 years 10 years 15 years Fig. 3. The effect of the investment amortization period on examination costs. Grey bars - CR. Black bars -chest CR unit. White bars - conventional plain-film. E LL INTEREST RATE Fig. 4. The effect of the investment interest rate on examination costs. Bar colours as in Fig. 3. cost of conventional imaging. When the amortization period was defined as 5 years, the cost of CR was 16% (FIM 6) higher than the cost of conventional radiography. Defining the amortization period as 10 years resulted in a cost difference of 9% (FIM 14) while 15 years resulted in a 6% difference (FIM 10). Changes in the interest rate also increased the cost of digital examinations more than the cost of conventional examinations (Fig. 4). At an interest rate of O%, the digital examination was 9% (FIM 14) more expensive than the conventional examin- Discussion The transition from a conventional film-screen radiology system to a digital filmless hospital requires careful planning. One plan would be to digitize one modality at a time, which would result in a transitional phase during which both hardand soft-copy images were produced simultaneously; this was done in our department. A second plan would be to make the whole hospital filmless overnight with no transitional period (1). In planning the equipment acquisition for the new Radiology Department in the Central Hospital of Vaasa, a decision was made to digitize the activity in stages, and to begin by introducing CR for most plain-film examinations. In 1994, the Radiology Department was in a transitional stage with plain radiography being done partially with CR and partially with conventional film-screen radiography. The interpretation and reporting was done mostly on film. Initial capital costs and film costs were higher in CR than in conventional radiography. We therefore decided to investigate whether it was possible at this stage to use the advantages of CR to create compensatory savings. The analyses listed in the literature used the traditional cost accounting method. We chose to perform our cost calculations with the ABC system because it enabled us to trace the different activities and thus gave a more precise allocation of the costs in the final products (3). All costs linked to a defined activity in the Radiology Department during the space of one year were taken into account in these cost analyses. Imaging was subdivided into various activities, making it possible to compare the effects of the digital and conventional techniques on the different functions in the department. The utilization rate of equipment was accounted for by allocating costs to functions according to operation times and frequencies. Because the image distribution net did not reach all hospital units, we could not analyse the potential effect of digitization on the costs of image transfers between the Radiology Department and other hospital units. According to earlier studies, these costs are significant (17). Replacing conventional radiography with a PACS system throughout the hospital has been considered cost effective because it reduces both the operating costs of the Radiology Department and the time patients stay in hospital (, 10, 16). 05

8 J. ALANEN ET AL. Digital radiology has thus been shown to pay for itself within years (5, 14). On the other hand, it has been argued that PACS does not reduce the running costs of a radiology department (1, 7, 15). CR has been recommended for bedside imaging because it gives a constant image quality. However, the use of phosphorus plates has not been shown to reduce examination costs in ICU patients (6, ). One reason for these differences in the various cost analyses is that cost factors are handled in a variety of ways. For example, equipment costs have been estimated both as decreasing by 50% within -5 years (1, ) and as remaining stable in the future (7). Maintenance cost estimates also vary: from 5% to 15%. These analyses use traditional cost accounting and are based partly on expert opinion regarding the ways in which digitization will affect operating costs. As our study used ABC accounting, it is based on actual operating costs. We have shown that in our department CR was 9% more expensive than conventional radiography. The difference was due to the higher initial capital cost of digital imaging. This cost was approximately twice as high as the initial capital cost of conventional film processing (Table 4). Our equipment was installed in 1993, at a time when the purchased machines were a novelty and thus more expensive than today. If we had calculated with a 30% lower equipment price, the cost of CR would have been reduced to almost that of conventional imaging. The effect of capital costs on total costs depends on the amortization period for capital goods and on the interest rate. Reducing the amortization time from 10 to 5 years increased the cost of digital imaging more than that of conventional imaging. Raising the interest rate had a similar effect. The actual lifetime of equipment varies according to use. Estimating lifetime is important in accounting the cost effects of equipment acquisition. In cost accounting, the amortization time affects the cost of expensive equipment more than it affects less expensive equipment. In the municipal health service in Finland, equipment is often acquired on accrued tax revenue. If digital equipment is not purchased, the capital saved is not usually consolidated. Therefore we decided to use an interest rate of 0% in this analysis and to compare our result with results obtained from interest rates on capital investment of 4% and %. Constant image quality has been considered an advantage in CR because it results in fewer retakes owing to faulty exposure (, 11). This reduces film costs, patient radiation doses, and examination times; at the same time, examination capacity increases (1 1). Despite the - to 3-fold price of laser film compared with conventional film, film costs were the same in our groups. This was largely due to the capability of combining several images on one film on the workstation and thus reducing film costs. The effect on film costs of the reduced number of retakes was, however, minor. According to an earlier (internal) study at the Central Hospital of Vaasa, approximately 6% of conventional filmscreen exposures required retakes. In our study, there were retakes in less than 1% of CR exposures. If all examinations had been performed with the conventional method, the added cost due to retakes would have been approximately FIM per year, i.e. approximately FIM 0.44 per plain radiograph. Images taken in ICU patients were sent to the ICU workstation, thus potentially accelerating the transfer of image information. However, this did not lead to any cost benefit as the ICU physicians still wanted the images on film so as to be able to compare them with earlier images: this was not possible on a workstation with one monitor. In our study, the average time used on the main procedure (the patient stay in the examination room) was the same for CR and film-screen radiography. The staff considered CR to be more pleasant and less demanding than conventional radiography. Therefore the time gained from a reduced number of retakes improved their working conditions, although not the capacity of the Radiology Department. However, the option of faster imaging may well result in a more obvious economic benefit in radiology departments that perform a greater number of examinations than our department. According to our results, the acquisition of CR equipment cannot be justified by cost benefits unless the acquisition leads to a complete PACS with image archives and an image distribution net throughout the hospital, thus drastically reducing the use of film. Step-by-step acquisition of CR with a planned transition to more extensive digitization can be justified by a better quality of examination during the transition period. According to the experiences of HRUBY (9), a PACS that covers the entire hospital pays for itself within 5 years. The cost benefit is seen as the result of the reduced film and archive costs as well as the result of a faster distribution of information which could reduce the length of a patient s stay in hospital. If all the advantages provided by PACS were utilized in the Central Hospital of Vaasa, the reduction in film costs would be about FIM per year, and in image sorting and traffic about FIM per year. In addition, the building of a new conventional film archive, which is already planned, would no longer be necessary. 06

9 COSTS OF PLAIN-FILM RADIOGRAPHY REFERENCES 1. ANDRIENSSEN J. H. T. H., TER HAAR ROMENY B. M., BAR- NEVELD BINKHUYSEN E H. & VAN DER HORST-BRUINSMA I. E.: Savings and costs of a picture archiving and communication system in the University Hospital Utrecht. Proc. SPIE Med. Imaging 1093 (199), 57.. BEARD D., PARRISH D. & STEVENSON D.: A cost analysis of film image management and four PACS systems. Proc. Med. Imaging 4 (1990), CHAN Y. C.: Improving hospital cost accounting with activity-based costing. Health Care Manage. Rev. 1 (1993), COOPER. & KAPLAN R. S.: How cost accounting systematically distorts product costs. In: Accounting and management. Field study perspectives. Edited by W. J. Bruns & R. S. Kaplan. Harvard Business School Press, Boston MA CYWINSKI J. K. & VAN DEN BRINK J. A,: Review Of experience with PACS cost analysis model. Proc. Med. Imaging 1093 (199), DON S., ALBERTINA M. J., AMMAN D. L., EVENS R. G. & SIEGEL M. J.: Soft-copy computed radiography in neonatal and pediatric intensive care units. Cost-savings analysis. Radiology 197 (1995), DREW F? G.: Cost analysis of image management and communications systems (IMACS). In: Digital imaging, Medical Physics monograph no., p Edited by W. R. Hendee & J. H. Trueblood. Medical Physics, Madison, WI GLAZER H. S., Mum E., SAGEL S. S. & JOST R. G.: New techniques in chest radiography. Radiol. Clin. North Am. 3 (1994), HRUBY W.: Das filmlose Krankenhaus - Realitat oder Illusion. ROFO 16 (1995), MOSSER H., URBAN M., PARTAN G. et a1 : Clinical rutine operation of a filmless radiology department. 0 months experience. Symp. Comput. Assist. Radiol., Carlsbad, p. 30. Symposia Foundation of Carlsbad, CA MURPHJZ M. D., QUALE J. L., MARTI N. L., BRAMBLE J. M., COOK L. T. & DWYER S. J. 111: Computed radiography in musculoskeletal imaging. State of the art. AJR 15 (199), NISSEN-MEYER S. A., FINK U., PLEIER M. & BECKER C.: The full-scale PACS archive. A prerequisite for the filmless hospital. Acta Radiol. 37 (1996), 3.. PETERS F? E., DYKSTRA D. A., WIE~MANN W., SCHLUCHTER- MANN J. & ADAM D.: Wissenschaftlichkeitsvergleich zwischen digitaler Luminisenzradiographie und konventioneller Filmverarbeitung in der Intensivmedizin. Radiologe 3 (199), VAN POPPEL B. M., VAN GENNIP E. M. s. J., BAKKER A. R. & WILMINK J. B. M.: First results with the software package capacity for cost modeling of PACS. Proc. Med. Imaging 4 (1990), SAARINEN A. O., HAYNOR D. R., LOOP J. W. et a].: Modeling the economics of PACS. What is important? Proc. Med. Imaging 1093 (199), SIEGEL E. L.: The transition to the filmless imaging department. Early experience at the Baltimore VA Hospital. Symp. Comput. Assist. Radiol., Carlsbad, p. 5. Symposia Foundation of Carlsbad, CA STRAUB W. H. & GUR D.: The hidden costs of delayed access to diagnostic imaging information. Impact on PACS implementation. AJR 155 (1990), STRICKLAND N. H.: Review article. Some cost-benefit considerations for PACS. A radiological perspective. Br. J. Radiol. 69 (1996), T~JCKER D. M., BARNES G. T. & KOEHLER. E.: Picture archiving communication systems in the intensive care unit. Radiology 196 (1995),