Secondary Logo

Journal Logo

Physician–Scientists in Japan: Attrition, Retention, and Implications for the Future

Koike, Soichi MD, PhD; Ide, Hiroo PhD; Kodama, Tomoko MD, PhD; Matsumoto, Shinya; Yasunaga, Hideo MD, PhD; Imamura, Tomoaki MD, PhD

doi: 10.1097/ACM.0b013e31824d47e8

Purpose To investigate career trends for physician–scientists in Japan.

Method The authors analyzed 1996–2008 biennial census survey data from Japan’s national physician registry to examine trends over time in the numbers and proportion of physician–scientists by sex and years since registration. They also analyzed the transition of registered physicians into and out of the physician–scientist field across two sets of two consecutive surveys (1996–1998 and 2006–2008).

Results The number of physician–scientists between 1996 and 2008 was stable, with a low of 4,893 and a high of 5,325. The number of younger physician–scientists (those registered 0–4 years at the time of the surveys) declined sharply, however, from 828 in 1996 to 253 in 2008. The number of female physician–scientists increased from 528 in 1996 to 746 in 2008. Across the two survey periods, about 30% of physician–scientists left the career path, but this attrition was offset by about the same number of new individuals entering the field.

Conclusions Although the total number of physician–scientists was relatively unchanged during the period studied, it is essential that educators and policy makers develop approaches to address underlying demographic changes to ensure an adequate age- and gender-balanced supply of physician–scientists in the future.

Dr. Koike is associate professor, Department of Planning, Information and Management, University of Tokyo Hospital, Tokyo, Japan.

Dr. Ide is assistant professor, Department of Planning, Information and Management, University of Tokyo Hospital, Tokyo, Japan.

Dr. Kodama is chief senior researcher, Department of International Health and Collaboration, National Institute of Public Health, Saitama, Japan.

Mr. Matsumoto is research fellow, Department of Planning, Information and Management, University of Tokyo Hospital, Tokyo, Japan.

Dr. Yasunaga is project associate professor, Department of Health Management and Policy, Graduate School of Medicine, University of Tokyo, Tokyo, Japan.

Dr. Imamura is professor, Department of Public Health, Health Management and Policy, Nara Medical University, Nara, Japan.

Supplemental digital content for this article is available at

Correspondence should be addressed to Dr. Koike, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan; telephone: +81 (3) 5800-8716; fax: +81 (3) 5800-8765; e-mail:

The physician–scientist career path, in which “individuals with an MD degree … perform medical research as their primary professional activity,”1 was originally promoted as allowing people with medical backgrounds to apply their knowledge and expertise to basic medical science. Decades after Wyngaarden2 first reported a shortage of physician–scientists in the United States, many young American physicians view the career path as lacking “professional security,” and those who leave the track cite career instability as a major reason.3

In Japan, as in the United States, there are concerns about a future shortage of physician–scientists. In 2010, four Japanese academic societies representing basic medical scientists jointly urged the government to take immediate action to resolve this crisis affecting both the medical science workforce and medical science research.4 The societies argued that the sharp decline in the number of medical school graduates aspiring to become physician–scientists could be linked to government policy changes, such as decisions to convert national universities into independent corporate entities, to reduce government subsidies, to introduce a mandatory two-year initial postgraduate clinical training period, and to cut the number of medical science posts available at national universities.4

Studies regarding the career paths of physician–scientists are limited in Japan, and, to the best of our knowledge, there have been no large-scale studies on the topic using data from the country’s national physician registry. Thus, we conducted a study in which we analyzed national physician registry data to investigate the number and proportion of physician–scientists relative to the total number of physicians and to track physicians’ transitions into different career paths, using both cross-sectional and longitudinal analyses. We present our findings here and consider their implications for the future.

Back to Top | Article Outline



As of 2008, Japan’s total population was 127.7 million people. Japan’s spending on health care was 8.1% of the gross domestic product(GDP) (compared with an average health expenditure of 8.9% of GDP for Organisation for Economic Co-operation and Development [OECD] countries). The proportion of physicians per 1,000 population was 2.1 (compared with an average of 3.1 per 1,000 for OECD countries).5

Japan has 79 national, prefectural, municipal, and private medical schools, 11 of which offered specialized MD-PhD programs as of 2006.6 Japan’s total medical school enrollment capacity was 8,846 students in 2010.

Japanese students enter medical school at the age of 18, following high school graduation. (Thirty-six medical schools [46%] offer programs for college graduates, but student quotas are limited.) After six years of undergraduate medical education, graduating medical students are eligible to take the national examination for physicians. On passing this exam, they are licensed, and they register as physicians with the Ministry of Health, Labour, and Welfare (MHLW).6

Since 2004, the government has required physicians to complete a two-year postgraduate clinical training program focusing on primary care and general medicine, regardless of their intended specialty, to see patients. This training, offered in designated university/university-affiliated (academic) hospitals or other training hospitals, must be completed before other postgraduate training (e.g., in a chosen specialty or graduate research program) may be undertaken.

Back to Top | Article Outline

Study sample

Under Japanese law, all physicians are required to report their status to the government every two years. This is accomplished through the MHLW’s biennial census survey, which collects data such as physician registration number, workplace name and address, type of work performed, and field of medicine. We obtained permission from the MHLW to use anonymized, selected data items for research purposes. Because of changes in the reporting format for the national physician registry over time, we only used data from the seven surveys conducted from 1996 through 2008. (Although the 2010 census survey was complete at the time of our study, 2008 data were the most recent data available to researchers.) We defined physician–scientists as registered physicians who indicated their “type of work” as (1) staff or graduate students at academic hospitals who were not practicing or (2) staff at educational or research institutions other than academic hospitals.

Back to Top | Article Outline


Across the seven surveys, we tracked the number and the proportion of physician–scientists relative to the total number of physicians according to sex and years since registration. For two sets of consecutive surveys (1996–1998 and 2006–2008), we also tracked shifts in the type of the work physicians reported conducting. We calculated retention rates to examine the dynamics of physician–scientists’ careers at the national level.

This study was approved by the research ethics committee of the Graduate School of Medicine and Faculty of Medicine, University of Tokyo.

Back to Top | Article Outline


Physician–scientists relative to all physicians in Japan

Between 1996 and 2008, the number of physician–scientists in Japan was relatively stable, with a low of 4,893 in 2008 and a high of 5,325 in 2000 (Table 1). However, within this same period, the total number of physicians increased from 248,070 to 286,699; this translated to a decline in the proportion of physicians who were physician–scientists, from 2.0% (5,041/248,070) in 1996 to 1.7% (4,893/286,699) in 2008.

Table 1

Table 1

Both the number of female physician–scientists and proportion of physician–scientists who were female increased between 1996 and 2008: There were 528 women among Japan’s 5,041 physician–scientists (10.5%) in 1996, whereas there were 746 women among the country’s 4,893 physician–scientists (15.2%) in 2008. During the same period of time, the total number of female physicians and the proportion of all physicians who were female increased from 33,229/248,070 (13.4%) to 51,997/286,699 (18.1%).

Back to Top | Article Outline

Physician–scientists by years since registration

From 1996 to 2008, the number of physician–scientists who had been registered physicians for 0 to 4 years decreased from 828 to 253. The number who had been registered for 5 to 9 years was 963 in 1996, peaked at 1,031 in 2000, and then decreased to 739 in 2008. The number who had been registered for 10 to 14 years was 878 in 1996, peaked at 889 in 1998, and fell to 611 in 2008. In comparison, the number of physician–scientists who had been registered for 20 to 24 years, 25 to 29 years, and at least 30 years increased steadily, from 525, 498, and 720, respectively, in 1996 to 893, 765, and 895, respectively, in 2008 (Figure 1).

Figure 1

Figure 1

Back to Top | Article Outline

Retention rate and career dynamics of physician–scientists

We analyzed the career transitions of physician–scientists over two periods of two consecutive surveys (1996–1998 and 2006–2008; see Supplemental Digital Figure 1,, panels A and B, respectively). The retention rate for physician–scientists was 69.7% (3,390/4,864) in 1996–1998 (Table 2) and 70.0% (3,221/4,603) in 2006–2008 (Table 3). Each of these time periods had an attrition rate of around 30%, but as individuals left the profession to pursue clinical career paths, about the same number of clinical physicians (1,595 in 1996–1998 and 1,301 in 2006–2008) entered the physician–scientist career path. This kept the size of the physician–scientist pool relatively stable across the two surveys within each period (4,864 in 1996 versus 4,985 in 1998; 4,603 in 2006 versus 4,522 in 2008). During both of these periods, the majority of physician–scientists came from or went to work at hospitals, and, in recent years, this number has been increasing.

Table 2

Table 2

Table 3

Table 3

When we examined the retention rates of physician–scientists by the number of years since registration (Tables 2 and 3), we found a lower retention rate for those who had been registered for 5 to 19 years in 2006–2008 than for those in these categories in 1996–1998. The retention rates were similar in both periods for physician–scientists who had been registered for 0 to 4 years (40.7% in 1996–1998 versus 38.4% in 2006–2008), but the actual number of individuals entering the physician–scientist career path fell sharply (690 in 1996–1998 versus 346 in 2006–2008). If this trend continues, there will be a decrease in the number of physician–scientists in the near future.

Back to Top | Article Outline


Although the total number of physician–scientists in Japan has remained stable over recent years, we observed two major trends within the physician–scientist workforce: a decrease in the number of young physicians entering the physician–scientist career path and an increase in the number of female physician–scientists. If the former trend continues, it is likely that Japan will face a shortage of physician–scientists in the future. The relative stability in the total number of physician–scientists during 1996–2008 can be attributed to the steady increase in the number of physician–scientists who have been registered for at least 20 years. This increase in the number of highly experienced physician–scientists is related to the creation of new medical schools in the 1970s7 and the accompanying increase in the number of medical school positions, which led to the formation of a cohort of physician–scientists with permanent positions. As this cohort of physician–scientists has aged, the age distribution of the physician–scientist workforce has become skewed. The skewed distribution will become more pronounced as the number of junior physician–scientists entering the profession decreases.

The mandatory initial postgraduate clinical training period introduced in 2004, which is facilitated by a physician-to-training-facility matching system, has led to a career pattern shift as many newly registered physicians move from academic hospitals affiliated with their medical schools to other training hospitals.8 Requiring all newly registered physicians to complete these two years of clinical training will automatically reduce the number of young physician–scientists by delaying young physicians’ entry into research careers. It may also discourage some from even considering the career path.

Ensuring that the number of physician–scientists and the quality of medical science research remain high is a major health policy issue in Japan, as in a number of other countries. Shimizu9 recently reported that the number of non-MD faculty members conducting basic medical sciences research in Japan in 2005 surpassed the number of MD faculty members doing such work and predicted that a serious shortage of medical educators was impending. He warned that this shortage will have a negative effect on the wider sphere of medical research—it will not be limited to the life sciences, translational research, or clinical research but will also affect the pharmaceutical and medical device industries and health policy research.

Previous studies in other countries1,10–12 have reported various approaches to counter the decreasing intake of physician–scientists, such as addressing financial and medical education issues and finding ways to increase the number of women entering the career path. We will consider each of these areas, below. The trends we found in this study—toward fewer junior physician–scientists and increasing numbers of female physician–scientists—are common in many countries, so we believe that most of these measures could be applied in Japan.

Country-specific context, however, should be taken into account when considering our results and possible steps. For example, in countries such as the United States, it has been argued that international medical graduates (IMGs) may represent a potential source of physician–scientists.13–15 The language and the licensing system in Japan, though, may discourage IMGs from practicing here.

Back to Top | Article Outline

Addressing financial issues: Tuition fees and research grants

In the United States, rising medical school tuition costs have been identified as a factor preventing some medical school graduates from entering the physician–scientist career track. The loan repayment programs that the National Institutes of Health introduced in 2002 fueled an increase in the number of individuals choosing research careers while in medical school or residency training.1 Providing centralized oversight and using institutional funds to offer financial support for physician–scientists’ career development may be a possible mechanism to catalyze growth in the number of funded physician–scientists at an institution.

In Japan in 2009, the average tuition fees for a first-year medical student at a national university were around ¥800,000 ($10,000), about the same as the average tuition fees for other first-year national university students. In contrast, the average tuition fees incurred by a first-year medical student enrolled at a private university were about ¥5 million ($62,500)—an amount higher than the average tuition fees of students in any other private university department (average ¥1.3 million or $16,400).16 Because half of Japan’s medical schools reside within national universities and have tuition fees equivalent to those of their universities’ other departments, prospective medical students who intend to seek careers as physician–scientists could reduce their tuition costs by choosing to attend national universities.

In view of such financial considerations, a career as a physician–scientist may not seem attractive to medical students or young physicians. It may be possible to attract more members of younger generations into the physician–scientist track by increasing the number and size of government-funded research grants, which could improve salaries and help decrease the burden of tuition fees (especially for graduates of private medical schools). Although government funding for life sciences research increased from ¥315.4 billion ($3.94 billion) in 2006 to ¥346.1 billion ($4.33 billion) in 2009 as a part of Japan’s third science and technology strategic plan,17 more government support is required.

Back to Top | Article Outline

Considering medical education options

Financial incentives are not, and should not be, the only tool used to entice new investigators into the field. Other approaches include informing potential medical school applicants about the other attractive features of the physician–scientist career path as well as providing support to those who are interested in pursuing the career path.

It is also important to increase younger generations’ awareness of the physician–scientist career path through undergraduate programs as well as by creating and/or supporting medical scientist training programs (i.e., MD/PhD programs) and research fellowships in subspecialties. Exposing medical students to research experiences can positively affect their career choices and attitudes about biomedical research.18 Recently, Japan’s government increased the national medical school enrollment capacity and allocated 17 of the additional seats for students in MD/PhD programs in fiscal year 2010. The number of seats allocated to physician–scientist training programs is expected to grow further, and there is some consideration of providing scholarships to medical students seeking careers as physician–scientists and of forming medical school consortia in fiscal year 2011.

Additionally, our study showed that 30% of physician–scientists left or entered the career path during 1996–1998 and 2006–2008, suggesting that transitions between physician–scientist and clinician careers are not rare. Thus, it is important to target and recruit young physicians with specific clinical experience who may be interested in the physician–scientist career pathway, because clinicians who participate in clinical research early in their careers are particularly successful in establishing careers as clinical investigators. Developing postgraduate programs in which they can develop the ability to transfer their clinical skills and knowledge to basic science and medical research is an option.19 In addition, “late bloomers”—individuals who have engaged in extended research training after completing medical school, residency, and clinical subspecialty training20 are in demand because these physician–scientists are able to successfully translate clinical practice issues into basic research questions.

Back to Top | Article Outline

Increasing the number of female physician–scientists

In the United States, the tendency toward a physician–scientist career path is relatively low, particularly among women.21 In the early 2000s, some researchers attributed the consistent rejection of the physician–scientist career path by women in their early 20s to difficulty balancing the demands of childbearing, family life, and career; competition with male colleagues; too little encouragement; and a lack of compelling role models.10

Although the proportion of women among physician–scientists in Japan is still low compared with that in other developed countries,22 the total number of female physicians has been increasing in Japan. Currently, the increase in the number of female physician–scientists is proportionate to the increase in the total number of female physicians being trained. It is important to find ways to encourage female medical students and physicians to pursue the physician–scientist career path.

Back to Top | Article Outline


There are several limitations to this study. These may be attributed to the study’s observational nature and the analysis of an existing database. Although the national physician registry data have the advantage of being census survey data, the registry does not collect qualitative or quantitative data on reasons for changing fields/specialties or information related to salary and academic position. There is also uncertainty as to whether missing data represent individuals who did not respond to the survey, who are working or studying abroad, or who are dead. Despite these limitations, our analysis of these nationwide data revealed the current status of physician–scientists in Japan and provides useful insights toward the study of the physician–scientist career pathway.

Back to Top | Article Outline


The number of physician–scientists in Japan between 1996 and 2008 was stable, but we observed two important demographic changes: Fewer junior physician–scientists entered the career path, but the number of female physician–scientists increased. Over the long term, fewer junior physician–scientists entering the career path will lead to a shortage in the number of physician–scientists and changes in the way they work. To ensure an age- and gender-balanced supply of physician–scientists in the future, measures to address the shortage should focus on the underlying determinants. Unless medical educators and policy makers take action, the shortage of physician–scientists will worsen, which will have implications for medical education, medical research, and associated industries.

Funding/Support: This study was conducted with support from the Health and Labour Sciences Research Grants (Research on Region Medical).

Other disclosures: None.

Ethical approval: The study was approved by the research ethics committee of the Graduate School of Medicine and Faculty of Medicine, University of Tokyo.

Previous presentations: Part of this work was presented at Association for the Study of Medical Education’s 2011 Annual Scientific Meeting in Edinburgh, United Kingdom.

Back to Top | Article Outline


1. Ley TJ, Rosenberg LE. The physician–scientist career pipeline in 2005: Build it, and they will come. JAMA. 2005;294:1343–1351
2. Wyngaarden JB. The clinical investigator as an endangered species. N Engl J Med. 1979;301:1254–1259
3. Donowitz M, Germino G, Cominelli F, Anderson JM. The attrition of young physician–scientists: Problems and potential solutions. Gastroenterology.. 2007;132:477–480
4. Physiological Society of Japan; Japanese Association of Anatomists; Japanese Biochemical Society; Japanese Pharmacological Society.. Petition to the Government and Ruling Party on Basic Medical Education and Activation of Medical Research [in Japanese]. Accessed February 20, 2011
5. Organisation for Economic Co-operation and Development.. OECD Health Data 2010 [CD ROM]. 2010. Paris, France OECD Publishing
6. Kozu T. Medical education in Japan. Acad Med. 2006;81:1069–1075
7. Ushiba D. Trends of medical education in Japan. Med Educ. 1985;19:258–265
8. Koike S, Ide H, Yasunaga H, Kodama T, Matsumoto S, Imamura T. Postgraduate training and career choices: An analysis of the National Physicians Survey in Japan. Med Educ. 2010;44:289–297
9. Shimizu T. Shortage of basic medical researcher—Current situation and countermeasures [in Japanese]. Paper presented at: Ministry of Education, Culture, Sports, Science and Technology, Japan; March 2011; Tokyo, Japan. Accessed January 22, 2012
10. Andrews NC. The other physician–scientist problem: Where have all the young girls gone? Nat Med. 2002;8:439–441
11. Ley TJ, Rosenberg LE. Removing career obstacles for young physician–scientists—Loan-repayment programs. N Engl J Med. 2002;346:368–372
12. Brass LF, Akabas MH, Burnley LD, Engman DM, Wiley CA, Andersen OS. Are MD-PhD programs meeting their goals? An analysis of career choices made by graduates of 24 MD-PhD programs. Acad Med. 2010;85:692–701
13. Vidyasagar D. Integrating international medical graduates into the physician–scientist pool: Solution to the problem of decreasing physician–scientists in the United States. J Investig Med. 2007;55:406–409
14. Puljak L. An overlooked source of physician–scientists. J Investig Med. 2007;55:402–405
15. Alexander H, Heinig SJ, Fang D, Dickler H, Korn D. Contributions of international medical graduates to US biomedical research: The experience of US medical schools. J Investig Med. 2007;55:410–414
16. Ministry of Education, Culture, Sports, Science and Technology.. FY 2009 Research Findings on First Year Average Student Payment to Private University and College [in Japanese]. Accessed May 19, 2011
17. Life Science Committee of Japan, Ministry of Education, Culture, Sports, Science and Technology, Japan.. Construction and Development of New Life Science in Japan—Interim Report [in Japanese]. Accessed May 19, 2011
18. Solomon SS, Tom SC, Pichert J, Wasserman D, Powers AC. Impact of medical student research in the development of physician–scientists. J Investig Med. 2003;51:149–156
19. Goldhamer ME, Cohen AP, Bates DW, et al. Protecting an endangered species: Training physicians to conduct clinical research. Acad Med. 2009;84:439–445
20. Varki A, Rosenberg LE. Emerging opportunities and career paths for the young physician–scientist. Nat Med. 2002;8:437–439
21. Guelich JM, Singer BH, Castro MC, Rosenberg LE. A gender gap in the next generation of physician–scientists: Medical student interest and participation in research. J Investig Med. 2002;50:412–418
22. Kodama T, Koike S, Matsumoto S, Ide H, Yasunaga H, Imamura T. The working status of Japanese female physicians by area of practice: Cohort analysis of taking leave, returning to work, and changing specialties from 1984 to 2004. Health PolicyIn press

Supplemental Digital Content

Back to Top | Article Outline
© 2012 Association of American Medical Colleges