The health care sector is an especially labor-intensive sector, and human resources are therefore important factors for the provision of health care to the general population. Nevertheless, medical treatment of a vast percentage of the German public might be at stake because Germany, like many other countries, faces a shortage of physicians in the near future.1–5 One main aspect of this potential lack of doctors is the demographic changes which Germany is undergoing. On one hand, there is a growing percentage of morbid and multimorbid persons needing medical treatment.6 This increasing number of diseased persons, on the other hand, is currently attended to by older doctors who will soon retire, and there is a limited number of young doctors to follow up and close the gaps.7
The shortage of young doctors is linked to an increasing number of medical school dropouts. In 2008, as many as 18% of German first-year students did not follow through to the final medical exams.7 Another reason for the upcoming shortage of physicians is the diversification of the labor market. An increasing number of medical graduates are leaving clinical practice and opting for noncurative medical jobs in other sectors such as health care management. Recently, the German Association of Statutory Health Insurance Physicians (SHI Physicians) reported that about 12% of medical graduates in Germany choose to work in areas other than patient care.7 Apart from that, the so-called feminization of medicine has also been discussed as a factor which adds to the physician shortage,8,9 as the percentage of female first-year students in medicine in Germany has increased to 63% in the year 2008. Although women doctors are increasing in number, they were simultaneously reported to be more likely to work part-time, and to be less likely to work longer hours.7 As female physicians were also reported to be slightly less satisfied with their jobs,10 they might think about leaving patient care more often than their male counterparts. Therefore, the effects supposedly associated with the feminization of medicine might even compound the pending lack of physicians.
In light of the interplay of these factors in the development and progression of the impending physician shortage, it is necessary to think about means to retain doctors in their curative jobs. One possible way to reach this goal is to create workplaces in clinical care that support the needs and wishes of doctors working there, and add to the physicians' job satisfaction.11–13 Therefore, this investigative study examined selected workplace characteristics and their association with junior doctors' considerations of leaving a job in patient care. (Junior doctors are those doctors in postgraduate training.) This approach seems to be sensible; for example, the theory of planned behavior claims that a “central determinant of behaviour is the individual's intention to perform the behaviour in question.”14 Based on the context outlined above, this study focused on which workplace-related factors might influence the retention of doctors in the patient care sector and could be addressed in advance in career advice or later on by departments and practices in order to retain human resources. Because of the so-called feminization of medicine, a gender-stratified approach was chosen.
Study design and participants
In cooperation with the Bavarian Medical Board Register, a participation letter and a questionnaire (September 2006) were mailed by the author to altogether 1,494 junior doctors who were registered as approbated doctors in Bavaria for the first time in 2005. The letter described the study contents and informed potential participants that their responses would be treated confidentially. The self-administered questionnaire included questions on demographic variables, medical education, the current workplace, and health-related questions. In a pilot test, 10 junior doctors of the Medical Faculty of Erlangen University filled in the questionnaire. For this study's analysis, which focuses on the question of which influenceable workplace factors make junior doctors think about leaving clinical practice, only young doctors working in patient care were included. By anonymously sending back the filled-in questionnaire, participating junior doctors provided informed consent. No ethical permission was necessary for this study.
Outcome variable: Thinking about giving up clinical practice
To identify the wish for leaving clinical practice, the questionnaire asked junior doctors to respond to the following statement: “During the last 12 months I thought about giving up clinical practice ....” Answers were given on a five-point response scale ranging from “never” to “every day.” For multivariate statistical analysis, these answers were dichotomized to junior doctors who never thought about giving up clinical care compared with junior doctors who thought about giving up clinical care at least a few times per year.
Predictor variables: Workplace factors
The following workplace factors were included as predictor variables in this analysis.
Junior doctors stated their specialty on a 13-point response scale (sorted from rather surgical to nonsurgical specialties). Internal medicine was regarded as the reference category. Specialties with fewer than 10 respondents (58 respondents) were excluded from multivariate analysis.
Support at work.
Junior doctors were asked about the general level of work-related support they receive from colleagues and superiors (dichotomized variable: a lot of or enough support versus little or no support).
Performance-related feedback from supervisors.
Feedback on performance from supervisors was surveyed according to the question “How often does your nearest superior talk with you about how well you carry out your work?” This feedback item was adopted from the Copenhagen Psychosocial Questionnaire (dichotomized variable: feedback always, often, or sometimes versus feedback seldom or never).15
Postgraduate training courses.
Participating doctors were also asked about the availability of postgraduate training courses (four-point response scale ranging from more than once a week to less than once a year).
Altogether, three factors regarding working time were considered as predictors for this analysis: overtime per week (three-point response scale, ranging from no overtime to more than 10 hours per week), night shifts per month (dichotomized variable: less than three night shifts versus three and more night shifts per month), and weekend duty (days per month) (dichotomized variable: zero or one weekend versus more than one weekend per month).
All analysis controlled for some personal factors, as they are generally considered as possible confounders: age (younger than 30 years versus older than 30 years), personal living conditions (living alone versus living with a partner), and children under the age of 15 living in the household (yes versus no).
All calculations were conducted using SPSS AMOS 18 (IBM, SPSS, Armonk, New York, 2010). Descriptive statistics for univariate categorical and continuous measures were generated. Frequencies for categorical variables and mean values with standard deviations (SDs) for continuous variables were calculated to describe the sample.
To evaluate the independent influence of workplace-related factors on thinking about giving up clinical practice for men and women, respectively, data were first stratified according to gender.
Chi-square tests were performed to evaluate differences between junior doctors who thought about leaving clinical practice and junior doctors who did not think about leaving clinical practice, with regard to all control and predictor variables for men and women. Additionally, Mann–Whitney U tests were used to evaluate differences between men and women with regard to control and outcome variables. Results are reported as P values, whereby a P value <.05 was regarded as statistically significant result.
Logistic regression analyses (backward selection) were conducted for men and women, respectively, to analyze the cross-sectional association between predictor variables and outcome variable. All predictor variables (workplace variables) were entered in the model at the same time and were excluded step by step according to their statistical significance (backward selection). Calculations were adjusted for control variables (age, personal living conditions, and living with children <15 years). Adjusted odds ratio (aOR) and corresponding 95% confidence interval (95% CI) were calculated.
Between September 2006 and March 2007, altogether 792 doctors anonymously sent back the filled-in questionnaire. Because of the preceding selection of participants by inviting only those firstly registered in Bavaria in 2005, the responding doctors were most likely to have had between one and a half to two years of work experience. Of the 792 respondents, 637 young junior doctors working in patient care met criteria for being included (55.8% women; response rate: 53%). Analysis of the questionnaires revealed that 16 (2.5%) of the respondents thought about leaving clinical practice every day, 50 (7.8%) a few times per week, 98 (15.4%) a few times per month, 169 (26.5%) a few times per year, and 304 (47.7%) never. Dichotomization of the outcome variable showed that 192 women (54.2%) thought about leaving clinical practice compared with 137 men (49.3%) (P = .216).
With regard to control variables, women were, on average, slightly younger than men. The mean age (SD) of all participants was 28.9 (2.4) years; women: 28.6 (2.5) years; men: 29.3 (2.2) years (P < .001). Altogether, 201 (56%) of the participating female doctors were living together with a partner compared with 170 (60.9%) of the participating male doctors (P = .286); 31 (11%) of the men as well as 39 (11%) of the women lived with children under 15 years (P = .856). Associations between control variables, predictor variables, and thinking about leaving clinical practice are shown in Tables 1 and 2.
Logistic regression analysis for predicting the influence of work-related factors on thinking about giving up clinical practice
The results of the logistic regression analysis support the hypothesis that different workplace factors predict thinking about giving up clinical practice by men and women, respectively (Table 3).
One important and gender-independent factor turned out to be work-related support from colleagues and superiors (men: aOR 2.70, 95% CI 1.20–6.04; women: aOR 3.46, 95% CI 1.67–7.17). In fact, lack of support was the strongest predictor for thinking about leaving clinical practice for women. Specialty also turned out to be of influence for both men and women. Junior doctors training in surgery thought less often about giving up clinical care compared with junior doctors training in internal medicine (men: aOR 0.22, 95% CI 0.09–0.51). Pediatrics and anesthesiology were two other specialties where women thought less often about giving up clinical care (anesthesiology: aOR 0.30, 95% CI 0.11–0.80; pediatrics: aOR 0.29, 95% CI 0.11–0.77). Fewer possibilities for postgraduate training were associated with thinking about giving up clinical practice for men (aOR 4.74, 95% CI 1.53–14.69). Also, more weekend duties were associated with thinking about giving up clinical care for male junior doctors (aOR 3.58, 95% CI 1.86–6.88), whereas women's thoughts about giving up clinical practice were associated with a lack of performance-related feedback from supervisors (aOR 1.88, 95% 1.23–3.14) and with overtime (aOR 1.82, 95% CI 1.02–3.25).
Discussion and Conclusions
To the authors' knowledge, this is the first study to provide a multivariate analysis regarding the possible association between thoughts about leaving clinical practice and workplace-related factors for junior doctors. Furthermore, it uses a gender-stratified approach to stay abreast of changes in the health care sector which might be associated with the so-called feminization of medicine.
The effect of workplace factors on junior doctors' thinking about leaving clinical practice was examined in order to identify means which prevent doctors from leaving clinical practice at an early stage. Previously, a large-scale questionnaire study of 5,461 participants from the Bavarian Medical Association reported that only a few of the young doctors were thinking about a career change compared with approximately 60% of the doctors in the age category 33–35 years,16 and another study of the German Medical Association and Association of the SHI Physicians showed that 55% of German doctors who choose to leave their jobs in patient care do so after their board certification as specialists.7 This study, however, determined that already 52.3% of the junior doctors with up to two years of job experience (mean age 29 years) had thought about leaving clinical practice at least a few times per year. Though it might need to be considered that especially dissatisfied young colleagues participated in this study, this number causes concern for the future and stresses the importance of “primary prevention” to retain dissatisfied junior doctors in the field of clinical practice at an early point of the medical career.
In general, the results of this study point out that work-related support, performance feedback from superiors, the availability of postgraduate training possibilities, overtime, weekend duty, and one's chosen specialty influence the frequency of how often junior doctors think about leaving clinical practice independently and in different ways. And though men and women did not differ with regard to thinking about giving up clinical care, the gender-stratified approach revealed that predictors differed for male and female doctors, with the exception of work-related support from colleagues and superiors, which turned out to be a gender-independent predictor. Other authors have also found support to be an important issue for junior doctors. When Brennan and colleagues17 interviewed junior doctors, they found the following topics to be of highest importance: newly gained responsibility, managing uncertainty, working in multiprofessional teams, experiencing the sudden death of patients, and feeling unsupported. That study's conclusion is that support and supervision should be made more readily available. But because of job enlargement in health care, support and the assurance of help seem nowadays not as readily available as they used to be, and junior doctors are often expected to take action without the support of more experienced colleagues.18 Furthermore, Joyce and colleagues19 have found a good support network to be associated with high job satisfaction.
Because the results of this study point out that a lack of support may also be a factor driving junior doctors away from clinical care, every clinical department should reflect on new approaches for supporting, educating, and training junior doctors in their first postgraduate years. This is even more important as additional unavailability of continuing postgraduate training and education seems to be a predictor for thinking about leaving the clinical field for men. Therefore, employers should not hesitate to support continuing education with time and money. Because this association has not yet been examined by other studies, future studies should control for the predictive value of the availability of postgraduate training possibilities.
Performance-related feedback by superiors is a significant predictor for women for thinking about leaving clinical practice. It has already been shown that women seem to be more insecure in estimating their own competencies; for instance, Lind and colleagues20 have reported that female students significantly underestimated their abilities in several competency domains, whereas male students accurately assessed or even overestimated their abilities. This difference, which is likely to last until becoming a junior doctor, may also have important implications for the provision of performance-related feedback in order to retain women in clinical practice.21 In fact, negative self-perceptions need to be addressed early on because they may affect career choice, performance, and persistence within those areas where the incompetencies are perceived.22,23 The lack of a role model of the same gender might also increase the effect of underestimating one's abilities. Murinson and colleagues24 have reported that “finding an exceptional role model” had tremendous impact on the emotional development of medical students, and Shuval and Adler25 found active identification (student emulates the role model) to be the most common student–physician interaction. Because women are often underrepresented in medical faculties (e.g., assistant, associate, and full professors26), female role models are very likely more scarce than male role models, especially in specialties such as surgery.27 This hampers the possibilities for women to emulate a role model of the same gender and might increase their wish for more performance-related feedback. Additionally, more female than male junior doctors reported receiving feedback only seldom or never in this study, which indicates a possible gender difference in the provision of feedback.
Doctors' chosen specialty, though not a factor which can easily be addressed as such in preventive efforts, was a statistically significant predictor for thoughts of leaving a clinical job for men and women. Especially surgeons were reported to be “born,” not “made,” as they usually decide very early on their specialty.28 Moreover, students or junior doctors aspiring to careers in surgery were also found to place the most emphasis on prestige compared with their colleagues who chose nonsurgical specialties.29,30 These findings support the notion that surgeons may be more likely to have certain personality traits which may make it easier for them to adapt to clinical practice. Thus, male junior doctors working in surgical specialties (compared with internal medicine) were less likely to think about leaving clinical practice in this study. Female junior doctors are more likely to choose nonsurgical specialties.31 This decision was found to be associated with an emphasis on lifestyle and not on prestige.29,30 Even when women have chosen to specialize in surgery, they were reported to be more concerned with family demands than men and to translate that priority into a greater commitment to the family.32 Hence, addressing underlying aspects of career choice such as prestige or lifestyle might help to retain junior doctors in patient care.
In accordance with the concern for family demands by women, overtime was a predictor for their thinking about leaving clinical practice. West and colleagues33 have found female internal medicine residents to be more likely to assign great importance to family time, a finding which might be due to gender differences in roles at home. With regard to men, overtime did not influence thinking about leaving clinical practice. This is mirrored in another evaluation of this study cohort, where a positive association between overtime and success could be detected, which supports the theory that junior doctors experience overtime not only as strain but also as reward.34 Nevertheless, a higher frequency of weekend duty was associated with thinking about leaving clinical care for men. Both results for men and women are supported by Dorsey and colleagues,35 who found an increasing wish for controllable lifestyles in male and female doctors over the last years. This analysis leads to the conclusion that this wish might have different outlets in men and women.
A limitation of the present study is the response rate of only 53%. Nevertheless, the response rate, although low, is higher than in other published questionnaire surveys, and also higher than the average for mail surveys, which often receive a 25% response rate.36,37 Though the present cross-sectional design of the study does not allow for analyzing cause–effect relations, the results of this study reveal a need for future prospective research, which should also address the association between thinking about giving up clinical practice and doing so.
The author wishes to thank Hans Drexler, MD, and Klaus Schmid, MD, University of Erlangen-Nuremburg, for the support of the study. Apart from that, she would like to thank Kathy Bischof for her instantaneous language support, and Ulrike Zier for the helpful statistical discussions.
This study was funded by the German Medical Association and technically supported by the Bavarian Medical Association.
Participants were informed about study contents and that their responses would be treated confidentially. All junior doctors provided informed consent. No ethical permission was necessary for this kind of study.
1 Sheldon GF, Ricketts TC, Charles A, King J, Fraher EP, Meyer A. The global health workforce shortage: Role of surgeons and other providers. Adv Surg. 2008;42:63–85.
2 Kane GC, Grever MR, Kennedy JI, et al. The anticipated physician shortage: Meeting the nation's need for physician services. Am J Med. 2009;122:1156–1162.
3 Toker A, Shvarts S, Glick S, Ruveni H. A report card on the physician work force: Israeli health care market—Past experience and future prospects. Health Policy. 2010;97:38–43.
4 Estryn-Behar M, Doppia MA, Guetarni K, et al. Emergency physicians accumulate more stress factors than other physicians: Results from the French SESMAT study. Emerg Med J. 2011;28:397–410.
5 Weiß J. The lack of general practitioners: How to overcome the crisis by providing structured professional training [in German]. Dtsch Med Wochenschr. 2010;135:26.
6 Kurth BM. Demographic changes and their consequences for the German health care system [in German]. Bundesgesundheitsbl Gesundheitsforsch Gesundheitsschutz. 2001;44:813–822.
7 Kopetsch T. German Health Care Runs Out of Doctors! [in German]. Berlin, Germany: German Medical Association and Association of the SHI Physicians; 2010.
8 McKinstry B. Are there too many female medical graduates? Yes. BMJ. 2008;336:748.
9 Dacre J. Are there too many female medical graduates? No. BMJ. 2008;336:749.
10 Bovier PA, Perneger TV. Predictors of work satisfaction among physicians. Eur J Public Health. 2003;13:299–305.
11 Janus K, Amelung VE, Gaitanides M, Schwartz FW. German physicians “on strike”—Shedding light on the roots of physician dissatisfaction. Health Policy. 2007;82:357–365.
12 Mohr DC, Burgess JF. Job characteristics and job satisfaction among physicians involved with research in the Veterans Health Administration. Acad Med. 2011;86:938–945.
13 Cox M, Kupersmith J, Jesse RL, Petzel RA. Commentary: Building human capital: Discovery, learning, and professional satisfaction. Acad Med. 2011;86:923–924.
14 Ajzen I, Fishbein M. Understanding Attitudes and Predicting Social Behavior. Engelwood Cliffs, NJ: Prentice-Hall; 1980.
15 Kristensen TS, Hannerz H, Høgh A, Borg V. The Copenhagen Psychosocial Questionnaire—A tool for the assessment and improvement of the psychosocial work environment. Scand J Work Environ Health. 2005;31:438–449.
16 Bornschein S, Erbas B, Borelli S, et al. Working time and job satisfaction of employed medical doctors in Munich [in German]. In: Schwartz FW, Angerer P, eds. Arbeitsbedingungen und Befinden von Ärztinnen und Ärzten. Cologne, Germany: Deutscher Ärzteverlag; 2010:65–77.
17 Brennan N, Corrigan O, Allard J, et al. The transition from medical student to junior doctor: Today's experiences of tomorrow's doctors. Med Educ. 2010;44:449–458.
18 Dudley HA. Stress in junior doctors: 1—Stress and support. BMJ. 1990;301:75–76.
19 Joyce CM, Schurer S, Scott A, Humphreys J, Kalb G. Australian doctors' satisfaction with their work: Results from the MABEL longitudinal survey of doctors. Med J Aust. 2011;194:30–33.
20 Lind DS, Rekkas S, Bui V, Lam T, Beierle E, Copeland EM 3rd. Competency-based student self-assessment on a surgery rotation. J Surg Res. 2002;105:31–34.
21 Minter RM, Gruppen LD, Napolitano KS, Gauger PG. Gender differences in the self-assessment of surgical residents. Am J Surg. 2005;189:647–650.
22 Beyer S. Gender differences in the accuracy of self-evaluations of performance. J Pers Soc Psychol. 1990;59:960–970.
23 Phillips D. The illusion of incompetence among academically competent children. Child Dev. 1984;55:2000–2016.
24 Murinson BB, Klick B, Haythornthwaite JA, Shochet R, Levine RB, Wright SM. Formative experiences of emerging physicians: Gauging the impact of events that occur during medical school. Acad Med. 2010;85:1331–1337.
25 Shuval JT, Adler I. The role of models in professional socialisation. Soc Sci Med. 1980;14A:5–14.
26 Wright S, Wong A, Newill C. The impact of role models on medical students. J Gen Intern Med. 1997;12:53–56.
27 Park J, Minor S, Taylor RA, Vikis E, Poenaru D. Why are women deterred from general surgery training? Am J Surg. 2005;190:141–146.
28 Thomas R. How do doctors choose their specialty: First love, arranged marriage or second time around? And how may an affair with MMC change this? Clin Med. 2008;8:490–492.
29 Glynn RW, Kerin MJ. Factors influencing medical students and junior doctors in choosing a career in surgery. Surgeon. 2010;8:187–191.
30 Azizzadeh A, McCollum CH, Miller CC 3rd, Holliday KM, Shilstone HC, Lucci A Jr. Factors influencing career choice among medical students interested in surgery. Curr Surg. 2003;60:210–213.
31 Lefevre JH, Roupret M, Kerneis S, Karila L. Career choices of medical students: A national survey of 1,780 students. Med Educ. 2010;44:603–612.
32 Yutzie JD, Shellito JL, Helmer SD, Chang FC. Gender differences in general surgical careers: Results of a post-residency survey. Am J Surg. 2005;190:978–983.
33 West CP, Drefahl MM, Popkave C, Kolars JC. Internal medicine resident self-report of factors associated with career decisions. J Gen Intern Med. 2009;24:946–949.
34 Ochsmann E, Lang J, Drexler H, Schmid K. Stress and recovery in junior doctors. Postgrad Med J. 2011;87:579–584.
35 Dorsey ER, Jarjoura D, Rutecki GW. The influence of controllable lifestyle and sex on the specialty choices of graduating U.S. medical students, 1996–2003. Acad Med. 2005;80:791–796.
36 Schmidt H, Vermeulen L, Van Der Molen H. Longterm effects of problem-based learning: Competencies acquired by graduates of a problem-based and a conventional medical school. Med Educ. 2006;40:562–567.
© 2012 Association of American Medical Colleges
37 Armstrong J, Lusk E. Return postage in mail surveys: A meta-analysis. Public Opin Q. 1987;51:233–248.