According to the Global Gender Gap Report 2014, Japan ranks 104th out of 136 countries in a comprehensive index of gender equality of health, education, economy, and politics, indicating that women’s social progress in this country lags far behind other nations. Women remain underrepresented in medicine in Japan, and Japan continues to rank last (i.e., 20% in 2014)1 among Organisation for Economic Co-operation and Development (OECD) countries in its percentage of female doctors—a consistent pattern for the last decade.2
Previous studies have established that female doctors in Japan are more likely than male doctors to resign from full-time practice at the time of life events related to gender role responsibilities including marriage, childbirth/rearing, and elderly nursing care.3,4 In a survey3 of 711 female doctors who graduated from two medical schools in Japan, about half had resigned from full-time employment because of difficulties in balancing work and child care responsibilities. Furthermore, the study reported that once those women resigned, only one-third returned to full-time employment.3 This gendered pattern of resignation limits women’s access to specialty training opportunities and the potential for pursuit of an academic career.
Indeed, in a previous study,4 we found that women were less likely than men to obtain a specialty board qualification (women vs. men, 59% vs. 84%) or doctor of medical science (DMSc) degree (women vs. men, 40% vs. 53%), which is an advanced degree beyond the degree necessary to practice medicine, similar to a British degree of doctor of philosophy in medicine. Because specialty board qualifications or DMSc degrees are essential for women to be promoted to higher positions in clinical practice or academia, respectively, the gender differences discussed above perpetuate strikingly substantial gender differences in leadership positions in Japan. For example, in 2011, women constituted only 2.6% of all full-time professors and only 2 out of 80 deans of medical schools,5 5.9% of all council members of medical societies,6 and only 3 of 29 executive board members of the Japan Medical Association.7
The scarcity of women in leadership positions in Japan may be partially explained by the so-called “slow pipeline phenomenon”—which posits that differences exist because few women have been in the field long enough to have achieved leadership positions; it is also possible that the pipeline is leaky—such that gender differences like those discussed above are causing differential attrition of women over time.8 However, the leaky pipeline phenomenon is insufficient to explain gender inequality in leadership positions in Japan because the proportion of women in leadership positions is substantially lower than expected from the physician gender ratio in the current leadership generation.1 Of note, one study documented that Japanese female doctors were more likely to work part-time rather than full-time if they had a strong perception of gender-based obstacles.9 In addition, Japan may lag behind Western countries in gender equity because of a lack of a monitoring system based on any nationally representative database. For example, benchmarking of gender equity in the United States is possible thanks to the periodic publication of reports from the Association of American Medical Colleges,10 and the United Kingdom periodically conducts a survey of staffing levels of medical clinical academics in British medical schools.11 These reports have stimulated a great deal of research in Western countries to understand and ultimately narrow gender gaps in these societies.12 Evidence quantifying gender differences and exploring their causes in traditionally male-dominated academic medicine is particularly needed in Japan.
In this study, we focused on women with children, who likely face the greatest difficulties balancing work and personal life in Japan’s strongly gender-structured society. We sought to understand the determinants of attainment of specialty board qualifications or DMSc degrees in this vulnerable population so that appropriate targeted interventions might ultimately be implemented as necessary to reduce undesired gender disparities overall. In Japan, after completing a 2-year residency of mandatory postgraduate medical education,13 doctors may choose either clinical practice or research, or both. A specialty board qualification typically takes 3 to 5 years to attain; a DMSc degree typically takes 4 years to complete. Thus, the earliest age at which one could attain specialty board qualification would be 29 to 31 years, and the earliest age by which one might attain a DMSc degree would be 30 years. A person can obtain both qualifications simultaneously. Hence, the purpose of this study was to investigate what personal and professional characteristics of female doctors who started a family were associated with career development—specifically, attainment of specialty board qualifications and DMSc degrees.
This cross-sectional study is based on alumnae surveys of 13 private medical schools in Japan conducted between June 1, 2009 and May 31, 2011 aiming to identify the challenges related to attainment of work–life balance. A total of 9,544 female doctors were recruited via postal mail. We sent a questionnaire about women’s support in medicine to 2,029 women who agreed to participate with written informed consent, and 1,684 women returned their questionnaires (response rate from those providing informed consent: 83%). After excluding 224 alumnae from 1 medical school because of incomplete data on specialty board acquisition, 235 alumnae who had no experience of having a child, 208 alumnae who did not answer whether or not they had a child, 57 alumnae who did not answer whether they had acquired a DMSc, and 1 person who answered that she had obtained a DMSc although she was still 27 years old at the time of survey (which seemed unlikely), 882 subjects remained in the analytic sample.
The study was approved by the institutional review board at Teikyo University School of Medicine (no. 08-107).
The original questionnaire collected data about baseline characteristics, working conditions, satisfaction and motivation at work, experience and perception of gender-based career obstacles, and menstruation and pregnancy outcome. The items evaluated in this study were marital status (single/married/divorced or widowed), presence and number of children, specialty, husband’s occupation (medical doctor vs. other), specialty board qualification (obtained vs. not obtained), DMSc degree (obtained vs. not obtained), age at the time of the first child’s birth, job change at the time of first pregnancy or first childbirth (resigned/changed workplace vs. stayed at the same place), child care support at home and at work place at the time of first childbirth, and age at the time of survey. We used these items because of previous literature showing that these variables impact women’s participation and advancement in medical careers.3,4 Child care support availability was measured based on the degree of satisfaction with child care support availability at home (i.e., presence of family or babysitter) and in the workplace (i.e., nursery places) by using a five-point Likert scale (from strongly satisfied to strongly dissatisfied). We grouped the response patterns into “satisfied,” neither “satisfied” nor “dissatisfied,” and “dissatisfied” with child care support.
Because the purpose of this study was to investigate what factors were associated with attainment of specialty board qualifications and DMSc degrees, we created four outcome groups: attainment of both specialty board qualifications and DMSc degrees, specialty board qualifications alone, DMSc degrees alone, and neither. We used the chi-square test for categorical variables and an analysis of variance for continuous variables to investigate differences among the four career attainment groups. Univariable and multivariable multinomial logistic regression models were applied to estimate odds ratios (ORs) for associations between each explanatory variable and the three career attainment categories (with a reference category of having attained neither qualification) along with 95% confidence intervals (CIs). We calculated P values based on a type 3 analysis of effect. To determine explanatory variables in multivariable analyses, stepwise model selection was used. Age at the time of survey was further categorized into tertile groups (25–39, 40–49, 50–). Referring to gender ratio of clinical departments in a national survey of Japanese physicians, dentists, and pharmacists,1 we further categorized specialty into five groups: internal medicine; pediatrics/obstetrics–gynecology (which is popular with female doctors but with higher levels of resignation at the time of life events such as childbirth/rearing); anesthesiology/emergency medicine (which may enable women to balance between work and life because of shift work); otorhinolaryngology/dermatology/ophthalmology/psychiatry (which are popular with female doctors and in which resignation or job changes may be less common because of less workload); and others, including specialties constituting less than 4% of the total number of female doctors. We conducted all analyses using SAS statistical software version 9.3 (SAS Inc., Cary, North Carolina) and set statistical significance at P < .05.
The average age of participants was 47.0 years ± 11.2 SD; other characteristics are shown in Table 1. In total, 296 of 882 women (34%) obtained both specialty board qualifications and DMSc degrees, 292 women (33%) obtained specialty board qualifications only, 83 women (9%) obtained DMSc degrees only, and 211 women (24%) obtained neither. Age at the time of the survey was youngest among those who obtained neither (P < .0001). Women who obtained neither were more likely to have married medical doctors (P = .018), and their average age at the time of first childbirth was youngest (P = .001). The proportion of women who resigned from their jobs at the time of first pregnancy or first childbirth was highest among women who obtained neither qualification, followed by women who obtained DMSc degrees only, women who obtained specialty board qualifications only, and women who obtained both (P = .001). The proportion of women who were not satisfied with child care support at work was highest among women who obtained DMSc degrees only (P = .01), and the proportion of those who were not satisfied with child care availability at home was highest among women who obtained neither (P = .06).
Table 2 shows the results of univariable multinomial logistic regression models for the three career attainment groups with a reference of “neither qualification.” Significant variables at P < .05 were age at the time of first childbirth (P = .001), resignation from job at the time of first pregnancy or first childbirth (P = .001), child care support at work at the time of first childbirth (P = .015), husband’s occupation (P = .02), specialty (P < .0001), and age at the time of survey (P < .0001).
Table 3 shows the results of a multivariable multinomial logistic regression model for the three career attainment groups compared with women who obtained neither specialty board qualifications nor DMSc degrees. Variables with significant global effect included age at the time of first childbirth (P < 0.0001); resignation from job at the time of first pregnancy or first childbirth (P < .0001); child care support availability at work at the time of first pregnancy (P = .010); specialty (P < .0001); and age at the time of survey (P < .0001). Age at first childbirth significantly distinguished those respondents in the specialty board qualification plus DMSc group, as well as those in the specialty-board-qualification-alone group and those in the DMSc-alone group, from those in the neither-qualification group. For each year increase in age, the OR of being in the group with both qualifications increased by 25% (OR 1.25, 95% CI: 1.16–1.34; P < .0001), the OR for specialty board alone increased by 17% (OR 1.17, 95% CI: 1.08–1.25; P < .0001), and the OR for DMSc alone increased by 10% (OR 1.10, 95% CI: 1.00–1.20; P = .049). Respondents who resigned after their first childbirth were significantly less likely to be in the specialty board qualification plus DMSc group, as well as in the specialty-board-qualification-alone group, rather than in the group with neither qualification. Those who resigned were 64% less likely (OR 0.36, 95% CI: 0.22–0.60; P < .0001) to be in the group with both qualifications and 56% less likely (OR 0.44, 95% CI: 0.27–0.74; P < .0001) to be in the group with only a specialty board qualification. Compared with women whose specialty was “internal medicine,” women whose specialty was either “pediatrics/obstetrics–gynecology,” “anesthesiology/emergency medicine,” “otorhinolaryngology/dermatology/ophthalmology/psychiatry,” or “others” were more likely to obtain specialty board qualifications only or both specialty board qualifications and DMSc degrees except for both qualifications among the “anesthesiology/emergency medicine” group. Women whose age at the time of investigation fell into the range of 25 to 39 years were less likely to have attained advanced qualifications.
This study examined what professional and personal characteristics of female medical doctors in Japan were associated with career development, defined as attainment of specialty board qualifications and DMSc degrees. Our findings complement and extend the insights present in the existing international literature on the subject of women’s participation in the Japanese medical profession.
Our study demonstrated positive associations between age at the time of the first child’s birth and attainment of advanced qualifications. Early pregnancy age likely decreases the likelihood of obtaining these qualifications by decreasing training opportunities because of the difficulty of balancing work and child rearing—a classic collision of biological and professional clocks. Indeed, younger age at the time of childbirth has been reported to have important career implications among female doctors internationally. For example, Willett and colleagues14 conducted a resident survey in the United States and asked about participants’ intention to have children during residency; they found that women intentionally postpone pregnancy because of perceived threats to their careers. We also found that the majority of a previous study sample agreed that female doctors should bear a child after specialty board acquisition because of the difficulty of balancing between training for specialty board attainment and personal life.15 However, despite the perception that the ideal plan would be to delay childbearing, the study15 demonstrated that some respondents actually bore their first child before acquisition of specialty board qualifications or DMSc degrees. Hence, considering that it is very difficult to adjust timing for pregnancy in women before specialty board attainment, we believe that it is more important to improve working conditions to allow female medical doctors to balance between training and gender-role responsibilities.
Our study demonstrated that resignation at the time of the first childbirth was negatively associated with attainment of a specialty board qualification alone or combination with a DMSc. According to OECD health data from 2013,2 Japan currently faces a severe physician shortage (i.e., 2.2 practicing physicians per 1,000 population vs. average 3.2 physicians per 1,000 population among OECD countries overall). As a result, hospital medical doctors in particular are chronically exposed to heavy workloads, long working hours, and frequent on-call duties.16 Because substitute labor is generally not hired to compensate for physicians out on leave, but substitutes are hired when a physician resigns, female physicians may hesitate to take maternity leaves and instead resign altogether from full-time labor despite legal allowances for parental leaves.17
Our previous study18 reported that women often underestimate their abilities in clinical practice settings. This lower level of confidence among women might reflect a phenomenon described as “stereotype threat,” created in a male-dominated culture of medicine where women internalize gendered expectations4 in ways that may make them vulnerable to giving up more easily when confronted with the challenges of balancing between work and personal life. In this regard, interventions to boost women’s self-efficacy through mentorship programs may be particularly important to mitigate or reverse the negative impact of stereotype threat.19 In turn, women might be retained within the workforce and take their legally entitled leaves rather than resigning, which our study suggests is associated with substantially lower likelihood of attaining advanced qualifications that are necessary for advancement if they do reenter academic medicine as a career.
Of further note, because young women bearing children are more likely to still be in training and work at teaching hospitals, such institutions should provide greater resources for child care, including on-site 24-hour child nurseries and sick-child care; introduce reduced-hours employment options as recommended by Child Care and Family Care Leave Law in Japan; and provide sources of practical advice for balancing between work and gender-role responsibilities. According to a report based on hospitalists which was conducted by the Japan Medical Association in 2009,20 only half of the respondents answered that their workplace had an on-site nursery, and among these, only one-fourth had a 24-hour nursery place. The shortage of nursery places has not much improved since then because of very few subsidies from the Ministry in Japan. Thus, Japanese physician–mothers have few options for child care. Nowadays, the number of nuclear families is increasing in Japan,21 and female doctors cannot rely on extended family to serve as caretakers. Because of a shortage in numbers of nurseries and the high cost of private babysitters, many female doctors have no choice but to quit their jobs to take care of their children by themselves. Although dissatisfaction with child care at work had differing directional associations with outcomes on multiple-variable analyses in this study that are difficult to explain, lack of on-site child care availability likely partially drives the resignation decision that was observed to be significant in our model. Dissatisfaction with child care at work was strikingly high in all groups and merits intervention.
Our findings also demonstrated that attainment of specialty board qualification as well as DMSc was more likely among those specializing in fields other than internal medicine. Specialties other than internal medicine may make it easier for women to balance between work and life, allowing workforce retention and attainment of specialty board qualifications and the DMSc degree. In our study, the majority of women had specialty board qualifications, but a much smaller percentage of women had DMSc degrees. This trend is consistent with a national resident survey22 that showed that only 36% of 3,838 second-year residents wished to obtain DMSc degrees, while 93% wished to obtain specialty board qualifications. Recently in Japan, it was noted that the contribution to the top general medicine journals by Japanese clinicians was very small.23 In this regard, it is vital to foster those physicians inclined toward research and to promote the attainment of DMSc degrees, not only to ensure that there is gender equity in career advancement but also to allow the best minds, regardless of gender, to contribute to essential medical advances.
This study demonstrated that older age at the time of survey was more likely to be associated with the attainment of specialty board qualification and DMSc degrees. This finding may be explained by two scenarios. One is that medical doctors who have longer career experience are more likely to attain these qualifications because those who are older are exposed to educational opportunity longer than younger women. Alternatively, those who are older may have more pioneering spirits (prompting their entry into a heavily male-dominated profession) that might also predispose toward attainment of such qualifications.17
Our study has several strengths. These include the ability to demonstrate that life-event-associated characteristics such as early pregnancy age or resignation at the time of first child’s birth were negatively associated with career development. To our knowledge, none of the previous studies of physicians in Japan has investigated the effect of resignations related to childbirth on overall career development. Thus, our findings may help stakeholders to develop concrete strategies for career development among women medical doctors who face the difficulty of pursuing career and child rearing. This study also has several limitations that need to be addressed. First, our samples are only those who graduated from private medical schools; inclusion of national or public medical school alumnae might yield different results. Second, there might be unmeasured confounding factors, including socioeconomic status, social support from supervisors or colleagues, or individual attitudes and motivations. Third, resignation as measured in this study was the experience of resignation at the time of first child’s birth; we did not have information about history of any return to work after resignation. Fourth, younger women in our sample who did not obtain any qualifications at the time of survey might go on to obtain such qualifications later in their future. Although we adjusted associations for age at the time of survey, our results might underestimate the absolute proportion of women who ultimately attain these qualifications. Fifth, because of the cross-sectional nature of this observational study, causal relationships may not necessarily be inferred. Thus, the results of our study require caution in interpretation.
In conclusion, our findings strongly suggest that improvements in the working environment in Japanese medicine are necessary to help women remain active in their careers and balance the dual pursuit of their career and gender-role responsibilities. Changes in institutional support to mitigate these challenges may ameliorate the gender disparities that persist in attainment of senior and leadership positions in the Japanese medical profession.
Acknowledgments: The authors would like to thank the alumni board members of 13 medical schools for their collaboration in conducting the sequential surveys; the staff in the Department of Hygiene and Public Health, Teikyo University School of Medicine, for their feedback and advice in developing the questionnaire; Ms. Megumi Yukawa for her assistance in collecting data; and Mr. Yu Nomura for his assistance in developing the dataset.
3. Izumi M, Nomura K, Higaki Y, et al. Gender role stereotype and poor working condition pose obstacles for female doctors to stay in full-time employment: Alumnae survey from two private medical schools in Japan. Tohoku J Exp Med. 2013;229:233–237.
4. Yasukawa K, Nomura K. The perception and experience of gender-based discrimination related to professional advancement among Japanese physicians. Tohoku J Exp Med. 2014;232:35–42.
5. Yasukawa K. Gender inequality in academic medicine in Japan. Med Teach. 2013;35:700.
6. Support Committee for Women Surgeons, Japanese Society of Surgeons [Nihon Gekagakkai Josei Gekai Shien Iinkai]. Report of support for women doctors by sectional meetings of all medical societies [Nihon Igakukai Bunkakai Niokeru Joseiishi Shienno Genjoni Kansuru Chousa Houkokusho]. https://www.jssoc.or.jp/other/info/info20120516.pdf
. Accessed April 19, 2016.
8. Ramakrishnan A, Sambuco D, Jagsi R. Women’s participation in the medical profession: Insights from experiences in Japan, Scandinavia, Russia, and Eastern Europe. J Womens Health (Larchmt). 2014;23:927–934.
9. Nomura K, Gohchi K. Impact of gender-based career obstacles on the working status of women physicians in Japan. Soc Sci Med. 2012;75:1612–1616.
12. Holliday EB, Jagsi R, Wilson LD, Choi M, Thomas CR Jr, Fuller CD. Gender differences in publication productivity, academic position, career duration, and funding among U.S. academic radiation oncology faculty. Acad Med. 2014;89:767–773.
14. Willett LL, Wellons MF, Hartig JR, et al. Do women residents delay childbearing due to perceived career threats? Acad Med. 2010;85:640–646.
15. Nomura K. Career design of women doctors [Josei Ishi No Career Design]. Hospital [Byouin]. 2013;76:441–445.
16. Nomura K, Inoue S, Yano E. The shortage of pediatrician workforce in rural areas of Japan. Tohoku J Exp Med. 2009;217:299–305.
17. Nomura K, Yamazaki Y, Gruppen LD, Horie S, Takeuchi M, Illing J. The difficulty of professional continuation among female doctors in Japan: A qualitative study of alumnae of 13 medical schools in Japan. BMJ Open. 2015;5:e005845.
18. Nomura K, Yano E, Fukui T. Gender differences in clinical confidence: A nationwide survey of resident physicians in Japan. Acad Med. 2010;85:647–653.
19. Strong EA, De Castro R, Sambuco D, et al. Work–life balance in academic medicine: Narratives of physician–researchers and their mentors. J Gen Intern Med. 2013;28:1596–1603.
22. Kimura T, Nomura K, Takahashi O, Aoki M, Yano E, Fukui T. Characteristics of second-year residents intending to become a primary care physician [Primary Care I Wo Shikousuru Ninenjikenshuui No Tokusei Ni Kansurukentou]. J Jpn Prim Care Assoc [Nihon Primary Care Rengogakkaishi]. 2012;35:6–11.
23. Fukui T, Takahashi O, Rahman M. Japanese representation in leading general medicine and basic science journals: A comparison of two decades. Tohoku J Exp Med. 2013;231:187–191.