Jpn J Radiol DOI 10.1007/s11604-015-0413-6

ORIGINAL ARTICLE

Current radiologist workload and the shortages in Japan: how many full‑time radiologists are required? Akihiro Nishie1 · Daisuke Kakihara1 · Takeshi Nojo2 · Katsumasa Nakamura1 · Sachio Kuribayashi3 · Masumi Kadoya4 · Kuni Ohtomo5 · Kazuro Sugimura6 · Hiroshi Honda1 

Received: 30 January 2015 / Accepted: 9 March 2015 © Japan Radiological Society 2015

Abstract  Purpose  To clarify the workload of certified radiologists and to estimate the current manpower shortages in Japan. Methods  We conducted a questionnaire survey for accredited training institutions. The contents included the radiologist employment pattern (full vs. part time), the number of computed tomography (CT) and magnetic resonance imaging (MRI) examinations and their radiology reports, the number of radiation therapy planning sessions, and the time per week spent for each work activity. We also used the hospital survey reports of Japan’s Ministry of Health, Labor, and Welfare in our analyses. Results  The estimated numbers of CT and MRI interpretation reports and radiation treatment plans that one fulltime radiologist could complete within 1 hospital day (8 h) were 19.9 and 1.22, respectively. To complete all CT and * Akihiro Nishie [email protected]‑u.ac.jp 1

Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, 3‑1‑1 Maidashi, Higashi‑ku, Fukuoka 812‑8582, Japan

2

Department of Radiology, New Tokyo Hospital, 1271 Wanagaya, Matsudo, Chiba 270‑2232, Japan

3

Department of Diagnostic Radiology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku‑ku, Tokyo 160‑8582, Japan

4

Department of Radiology, Shinshu University School of Medicine, 3‑1‑1 Asahi, Matsumoto 390‑8621, Japan

5

Department of Radiology, Graduate School of Medicine, The University of Tokyo, 7‑3‑1 Hongo, Bunkyo‑ku, Tokyo 113‑8655, Japan

6

Department of Radiology, Kobe University Graduate School of Medicine, 7‑5‑2 Kusunoki‑cho, Chuo‑ku, Kobe, Hyogo 650‑0017, Japan









MRI reports, at least 2.09 times more full-time diagnostic radiologists are needed in Japan. For radiation therapy, at least 1.23 times more full-time radiation oncologists are necessary at large- and medium-scale hospitals, although the number of radiation oncologists needed in Japan is balanced to the current number. Conclusion  The number of full-time certified diagnostic radiologists for CT and MRI interpretation in Japan is insufficient. Centralized radiation therapy facilities may be more efficient for meeting the increasing demand. Keywords  Full-time certified radiologist · CT · MRI · Radiation therapy · Japan

Introduction Radiology services, including diagnostic radiology and radiation oncology, constitute an important component of a hospital’s clinical service. However, the rapid progress of imaging technology and innovative treatment methods has resulted in an increased workload for radiologists. Particularly, a large number of computed tomography (CT) and MR imaging scanners exist in Japan. According to a survey from 2003 to 2004, only 16.3 % of institutions operating MR imaging scanners had actually hired certified radiologists as staff members [1]. Official interpretation of imaging studies and radiation therapy planning should be performed by specialists, namely adequately trained, specialty-certified diagnostic radiologists and radiation oncologists, to maintain the quality of clinical services. Currently, it is known that there is a short supply of radiologists, but basic data on radiologist manpower still do not exist. In order to estimate the current manpower shortages, we administered a questionnaire survey to the accredited

13



Jpn J Radiol

Table 1  Summary of diagnostic radiology parameters regarding CT and MRI Parameter

CT

MRI

Total number of examinations Total number of reports with scanned regions provided Image interpretation rate Number of interpretation reports from full-time diagnostic radiologists

8,446,036 8,763,915 81.6 % 6,814,230

3,219,649 3,833,420 80.3 % 2,938,461

Number of current interpretation reports per day per one diagnostic radiologist Average work period of full-time diagnostic radiologists Time spent on image interpretation Number of interpretation reports per day per diagnostic radiologist within 1 hospital day (8 h) Number of interpretation reports per day per diagnostic radiologist within 1 hospital day if all the working hours are designated to interpretation Number of full-time diagnostic radiologists required to complete all CT and MRI interpretation reports in our survey Total number of examinations in September 2011 in Japan Estimated number of examinations with scanned regions considered per day in Japan (in 2011) Number of diagnostic radiologists registered in 2011

15.99 45.96 h/week 24.11 h/week 13.92 26.5

6.89

Number of full-time diagnostic radiologists required to complete all CT and MRI interpretation reports in Japan

9503 (2.09 times)

6.00 11.4

2605.1 (1.49 times) 2,357,580 122,316 4551

1,121,831 66,784

Data in parentheses are ratios of full-time diagnostic radiologists required to current full-time diagnostic radiologists

training institutions that play an important role in supplying certified specialists.

Methods The questionnaire survey was conducted in 2013 under the initiative of the future planning committee, specialty certifying committee (currently dealing with the function of the specialty certifying board and the Accreditation Board of Training Institutions), and the President of the Japan Radiological Society. The clinical data between April 2011 and March 2012 were collected. We selected 685 training institutions consisting of 190 training center hospitals, 487 regular training hospitals, and 8 specialty training hospitals that had been accredited as training institutions by the Society. All the hospitals employed at least one certified radiologist (range 1–55). In this article, the term radiologist refers to both diagnostic and radiation oncologists. The first portion of the questionnaire included the number, employment patterns (full vs. part time) as well as the radiologists’ gender, the total number, use of contrast medium, and region scanned in CT and MR imaging, the total number of radiology reports (by full- and part-time radiologists), and the total number of radiation therapy planning sessions, including external beam radiation therapy, brachytherapy, and ion beam radiotherapy, and radiologists’ workload. The second portion of the questionnaire consisted of the average time per week spent on the following working activities: interpretation of radiology studies,

13

planning of diagnostic studies, angiography and interventional radiology procedures, gastrointestinal fluoroscopy, endoscopy, ultrasound, nuclear medicine studies, radiation therapy, outpatient care, inpatient care, other clinical studies, and management. Data analyses were performed at the Department of Clinical Radiology, Kyushu University by two radiologists (A.N. and D.K.) under the mentorship of the chairman of the Future Planning Committee (H.H.). The parameters analyzed were the number of CT and MR imagings used for interpretation and radiation therapy planning within the regular hospital hours per day to be completed by one fulltime radiologist.

Results Diagnostic radiology The total number of certified diagnostic radiologists at the 422 institutions that responded to the questionnaire survey was 2362. Among them, 2308 radiologists responded concerning the employment status: 1714 full time (74.3 %) and 594 part time (25.7 %). For the radiologists with unknown status, we estimated the number by using the ratio of known full- and part-time radiologists, and we classified them as 1754 full- and 608 part-time diagnostic radiologists. Regarding gender, of 2300 radiologists, 1829 (79.5 %) were males and 471 (20.5 %) females. The main diagnostic radiology parameters regarding CT and MR imaging are summarized in Table 1.

Jpn J Radiol Table 2  Summary of regions scanned by CT and MRI CT

Ratio (%) MRI

Ratio (%)

Head and neck Body  One region  Two regions

29.2 65.2 26.5 18.9

Brain Head and neck MRA Neck Chest

37.3 21.6 2.2 1.7

 Three or more regions 19.8 Musculoskeletal 4.2 Cardiac 1.4

Abdomen Pelvis Large vessel Spine Musculoskeletal Cardiac

8.7 6.8 0.5 14.6 6.4 0.4

Total

Total

100.2

100

“One region” in body CT represents the chest, upper abdomen, or pelvis, whereas “two regions” represents the chest and upper abdomen or the upper abdomen and pelvis. “Three or more regions” represents chest to pelvis or the neck to pelvis

CT interpretation workload

Table 3  Average workloads and the details for full-time certified radiologists per week Items

Diagnostic radiology (h)

Radiation oncology (h)

Interpretation of radiology studies Planning of diagnostic studies Angiography or IVR procedures Gastrointestinal fluoroscopy Endoscopy Ultrasound Nuclear medicine studies Radiation therapy Outpatient care

24.11 (52.5 %) 1.39 (3.0 %) 3.87 (8.4 %) 0.53 (1.2 %) 0.23 (0.5 %) 1.04 (2.3 %) 3.44 (7.5 %) 0.57 (1.2 %) 1.17 (2.5 %)

1.07 (2.3 %) 0.2 (0.4 %) 0.08 (0.2 %) 0.08 (0.2 %) 0.01 (0.0 %) 0.07 (0.2 %) 0.23 (0.5 %) 23.20 (50.4 %) 10.00 (21.7 %)

Inpatient care Other clinical studies Management

0.46 (1.0 %) 2.09 (4.5 %) 7.06 (15.4 %)

2.10 (4.6 %) 2.28 (4.9 %) 6.75 (14.7 %)

Total

45.96 (100 %)

46.07 (100.1 %)

The first data given are the actual time spent (in hours), and the data in parentheses are percentages. Interpretation of radiology studies includes interpretation of CT, MRI, and plain films

The total number of patients who underwent CT was 8,446,036 according to the survey subjects. The study numbers per scanned region are summarized in Table 2. As for the study numbers, more than one region per examination was not considered. However, to determine the radiologist workload, the number of studies needed to be adjusted to the number of study regions, which was 8,763,915. The number of CT reports produced by radiologists was 7,155,090; the number of reports made by full-time radiologists was 6,814,230 and that by part-time radiologists was 340,860. The rate of studies reported by radiologists was 81.6 %. Since the number of workdays from April 2011 to March 2012 was 243 days, excluding Saturdays, Sundays, national holidays, and New Year’s holidays, one full-time radiologist completed 15.99 studies (Table 1) and one part-time radiologist completed 2.31 studies per day.

IVR interventional radiology

MR imaging interpretation workload

Shortages of certified radiologists

The total number of patients who underwent MR imaging was 3,219,649, and the corrected number of studies, similar to CT, was 3,833,420. Regions scanned are summarized in Table 2. The corrected total number of reports was 3,076,915; the number of reports by full-time radiologists was 2,938,461 and that by part-time radiologists was 138,454. The rate of interpretation by radiologists was 80.3 %. The daily workload of the full- and part-time radiologists was 6.89 and 0.94, respectively (Table 1).

We estimated the number of radiologists necessary to report all CT and MR imaging examinations. The total number of CT and MR imaging studies per year in our study population was 12,597,335. This was for 243 hospital days in 2013, and one radiologist could complete 19.9 studies within 1 hospital day. Therefore, the number of full-time radiologists needed was 12,597,335/243/19.9 = 2605.1. This result suggests that at least 1.49 times (2605.1/1754) more radiologists are actually needed.

Overall workload Diagnostic radiologists’ workloads and their details are shown in Table 3 and Fig. 1a. We estimated the number of CT and MR imagings to be interpreted by one full-time radiologist per 1 hospital day (8 h). Diagnostic radiologists actually worked 45.96 h/week. Therefore, one fulltime radiologist may actually complete (15.99 + 6.89) ×  40/45.96 = 19.9 studies within 1 hospital day (CT: 13.92, MR imaging: 6.00). Diagnostic radiologists were found to spend 52.5 % of their time for interpretation (Table 3); therefore, (13.92 + 6.00)/0.525 = 37.9 studies (CT: 26.5, MR imaging: 11.4) were the average number completed if all the working hours were designated to interpretation.

13



Jpn J Radiol

Fig. 1  The average percentage of workload and work activities among full-time diagnostic radiologists (a) and radiation oncologists (b). a The workload for endoscopy, radiation therapy, outpatient care, inpatient care, and other clinical studies was assembled as ‘other work.’ ‘Interpretation of radiology studies’ includes interpretation of CT, MRI, and plain films. IVR interventional radiology. b The workload of planning of diagnostic studies, angiography or IVR procedures, gastrointestinal fluoroscopy, endoscopy, ultrasound, nuclear medicine studies, and other clinical studies was assembled as ‘other work’

Estimation of nationwide shortages According to the report by the Ministry of Health, Labor, and Welfare of the Government of Japan, the number of CT and MR imaging studies performed in Japan in September 2011 was 2,357,580 for CT (at 11,397 healthcare facilities) and 1,121,831 for MR (at 5210 facilities) [2]. This estimation did not take the number of scanned regions into consideration. If the total number of studies estimated by the data on the study regions in our subject was applied, it was 2,446,311 (2,356,580 × 8, 763,915/8,446,036) CT and 1,335,689 (1,121,831 × 3 ,833,420/3,219,649) MR imaging studies. The number of workdays in September 2011 was 20; thus, the number of CT and MR imaging studies per day in Japan was 122,316 for CT and 66,784 for MRI. Therefore, 9503 [(122,316  + 66,784)/19.9] full-time radiologists were

13

required. Since the number of certified diagnostic radiologists in 2011 was 4551 according to the database of the Japan Radiological Society, at least 2.09 (9503/4551) times more full-time radiologists are needed. Radiation oncology According to our survey, the total number of radiation oncologists was 732. Among them, 664 responded in regard to their employment status: 497 were full time (74.8 %), 167 part time, and 68 unknown. We classified oncologists of unknown status as full and part time according to the ratio of oncologists with known status, and we included 548 (497 + 51) full-time and 184 (167 + 17) parttime oncologists: 573 males (87.2 %), 85 females (12.8 %), and 1 unknown. The main parameters related to radiation oncology are summarized in Table 4.

Jpn J Radiol Table 4  Summary of radiation oncology regarding treatment planning Total treatment plans Average work period of full-time radiation oncologists Time spent on treatment planning Treatment plan number per day per one full-time radiation oncologist Treatment plan number per day per one full-time radiation oncologist within 1 hospital day (8 h) Full-time radiation oncologists required to complete all treatment plans in our survey Total treatment plans in September 2011 in Japan Estimated total treatment plans per day in Japan (in 2011) Radiation oncologists registered in 2011

200,012 46.07 h/week 23.20 h/week 1.50 1.22 674.7 (1.23 times) 21,930 1096.5 923

Full-time radiation oncologists required to complete all treatment plans in Japan

898.8

A datum in parentheses is the ratio of full-time radiation oncologists required to current full-time radiation oncologists

Table 5  Summary of treatment plans completed by radiation oncologists Radiation therapy

Number of planning sessions

Ratio (%)

External beam radiation therapy  IMRT  Intraoperative  SRT  TBI Brachytherapy  Intracavitary  Implant Ion beam radiotherapy

188,818 7968 108 7485 1787 9770 6450 3320 1427

94.4 4.0 0.1 3.7 0.9 4.9 3.2 1.7 0.7

Total

200,012

100

IMRT, SRT, and TBI represent intensity-modulated radiation therapy, stereotactic radiotherapy, and total body irradiation, respectively

Radiation therapy planning The number of radiation treatment planning sessions is summarized in Table 5. The total number of planning sessions was 200,012. The percentages of IMRT and SRT were 4.0 and 3.7 %, respectively. In our survey of treatment planning, we did not discriminate between full- and parttime oncologists, and we assumed that no different times per treatment planning existed. Oncologist workload Oncologists’ workload and its details are shown in Table 3 and Fig. 1b. The average work hours for radiation oncologists was 46.07 h/week (excluding lunch time). The average oncologist spent 23.20 h (50.4 %) in actual treatment planning (Table 3). The average working hours of part-time oncologists was 4.72 h/week. Therefore, the total working hours for radiation therapy by all radiation oncologists per year was 660,089 (23.20 × 548 + 4.72 × 184) × 243/5 h.

The time needed per one treatment planning session including external beam radiation therapy, brachytherapy, and their related work was 3.30 h (660,098/200,012). The number of treatment plannings that one fulltime oncologist could complete per day (8 h) was 1.22 (8 × 0.504/3.30). Thus, one full-time radiation oncologist could complete 1.22 treatment plannings per day together with other clinical and non-clinical duties, and, if one could concentrate purely on treatment planning, one oncologist could complete 2.42 (8/3.30) plannings per day. Manpower of certified oncologists We estimated the supply status of oncologists by the total number of treatment plannings (200,012) divided by treatment plannings per one oncologist per year (1.22 × 243). The result was 674.7 full-time radiation oncologists needed to complete all treatment plannings while engaging in other clinical and non-clinical duties, and this was at least 1.23 times (647.7/548) the number of oncologists needed. The nationwide supply of radiation oncologist was estimated. Since the total number of treatment plannings in September 2011 was 21,930 (at 718 facilities) according to the Ministry of Health, Labor, and Welfare of the Government of Japan [2] and the number of workdays in this specific month was 20, the number of plannings per day was 1096.5. Therefore, 898.8 (1096.5/1.22) oncologists were needed. Since the number of certified oncologists in the database of the Japan Radiological Society is 923, our estimation was almost identical to the real number of radiation oncologists in Japan.

Discussion We performed a survey to determine the status of radiologists’ workload. The healthcare institutions included in our survey were large and medium-sized hospitals accredited

13



as training institutions for specialty certification, because such hospitals had hired more than one full-time radiologist and such radiologists’ performance probably reflected the average radiologist workload. The rate of interpretation of CT and MR imaging studies performed by radiologists was 81.6 and 80.3 %, respectively, and approximately 20 % was not interpreted by radiologists even in large- and medium-sized hospitals in our survey. Because of the small number of full-time radiologists in the small-sized hospitals (with fewer than 100 beds) [1, 3], the rate of studies interpreted by radiologists was probably much less. Surveys including largeand small-sized hospitals by Nojo et al. [3] and Imamura et al. [4] reported that only 40–80 % of CT and MR imagings were interpreted by radiologists. These studies suggested that the actual number of imaging studies was much higher and the shortage of diagnostic radiologists was even more serious. Other factors related to this situation were the wider CT and MR imaging scanning range due to technological advances, lower management costs, lower actual study costs, and decreased time for medical care [1, 5, 6]. Our literature review revealed the highest number (6130) of CT and MR imagings per year per radiologist in Japan, and it was 4.3 times higher than the average of 16 other countries including Germany, the USA, and Canada; on the other hand, the number of radiologist per 1 million population was the lowest (36) [1]. Combining their results and the data from the Government of Japan, at least 2.09 times the current number of full-time diagnostic radiologists is needed in Japan, 1.49 times more is needed in the largeand medium-sized hospitals, and even more radiologists are needed in small-sized hospitals. Our data also indicated that 37.9 studies (26.5 CT and 11.4 MR imaging examinations) could be completed by one full-time radiologist per 1 hospital day if radiologists could concentrate on interpretation of images. These figures should contribute to future planning of the manpower supply and help to solve the nationwide shortage of certified radiologists. For radiation therapy, our data suggest the shortages of radiation oncologists and that radiation oncologists worked after regular hospital hours to complete treatment planning. In fact, at least 1.23 times of the current number of radiation oncologists is needed in large- and medium-sized hospitals. On the other hand, when we consider the overall supply, the number of treatment plannings and of radiation oncologists was almost identical. One of the reasons for this result may be that almost all of the radiation therapy plannings was done in large- and medium-sized hospitals by certified radiation oncologists. Radiation therapy tended to be concentrated in relatively large cancer hospitals, not in small-sized hospitals. We predict that the treatment planning or performing actual treatment will take more time

13

Jpn J Radiol

because of the increased demand for radiation therapy due to the increased number of cancer patients in the elderly population and increased number of more sophisticated treatment methods including IMRT and SRT. Therefore, more full-time radiation oncologists will be required; on the other hand, centralized radiation therapy facilities may be more efficient to meet the increasing demand. The number of treatment plannings per 1 full-time radiation oncologist per day, 2.42, is an important piece of information for the future manpower supply. There are a few limitations to our study. First, the contribution of non-certified radiologists was not taken into consideration. Because radiologists in training usually make preliminary reports and carry out radiation therapy planning and technical work, such as injecting contrast medium, they significantly support the workload of certified radiologists. The time for the education of the certified radiologists was not considered. Second, other radiologist duties, such as interpretation of radiography, nuclear studies, interventional radiology, and other clinical duties, were not considered. Third, regional differences were not taken into consideration. To the best of our knowledge, this is the first report dealing with radiologists’ activities in Japan, where average physicians spent 60 % of their working time on actual medical care [1]. At least 2.09 times more diagnostic radiologists are needed to cover CT and MR imaging examinations in Japan, and even more radiologists are needed to cover other studies. The number of radiation oncologists is balanced to the actual number of radiation therapies nationwide, but to meet the increased demand for radiation therapy due to the increasing number of cancer patients in the elderly population and development of more sophisticated technology, more radiation oncologists are required. Acknowledgments  We thank Dr. Keiko Imamura, Department of Radiology, St. Marianna University School of Medicine, for providing advice on the questionnaire format. We also thank Dr. Yuichi Imanaka, Department of Healthcare Economics and Quality Management, Kyoto University Graduate School of Medicine, for advice on how to summarize the data. Conflict of interest  Dr. Ohtomo received a research grant from Daiichi-Sankyo Co. The other authors have no conflict of interest.

References 1. Nakajima Y, Yamada K, Imamura K, Kobayashi K. Radiologist supply and workload: international comparison—Working Group of Japanese College of Radiology. Radiat Med. 2008;26:455–65. 2. Overview of healthcare center (static and dynamic) survey and hospital report in 2011, from Ministry of Health, Labour and Welfare. http://www.mhlw.go.jp/toukei/saikin/hw/iryosd/11/ (in Japanese).

Jpn J Radiol 3. Nojo T, Sekimoto M, Imanaka Y, Ohtomo K. Research report concerning actual condition survey in imaging diagnosis of CT and MRI. Japan Radiological Society. 2010. http://www.radiology.jp/modules/news/article.php?storyid=840. Accessed 22 June 2010 (in Japanese). 4. Imamura K, Nakajima Y, Koba R, Kobayashi K, Kawabuchi K, Mizunuma K, et al. MRI in Japan: survey of system utilization. Jpn J Magn Reson Med. 2005;25:154–64 (in Japanese).

5. Imamura K. Medical Imaging: its medical economics and recent situations in Japan. Jpn J Med phys. 2006;26:85–96 (in Japanese). 6. Katada K. CT radiation exposure in Japan: analysis of social factors. Jpn J Radiol Technol. 2006;62:649–56 (in Japanese).

13