The medical specialties chosen by doctors for their careers play an important role in the development of health-care services. There are some recent studies of how different medical specialities are perceived or how choices are made [1–3]. Several determinants have been identified: the feminisation of medicine, lifestyle, specialist status and the prospect of future income, as well as the structural conditions of the various residency programmes. Most studies addressed medical students, not residents. As far as we know there appear to be only a few studies on personality traits, career motivation and personal life goals as influencing factors on speciality choice [4].

Residency programmes

The application and selection procedures for residency programmes, the length, quality and structure of the programme, work schedules, mentorship, annual vacations are also factors which are considered when choosing a speciality [18, 19, 24, 27, 28].

Most studies investigated only one or two of the factors identified as influencing speciality choice. One has to consider, however, that Swiss studies addressing issues of speciality choice are lacking to date.

The aims of this study were to investigate (1) the development of the residents' speciality choices since graduating from medical school, and the differences compared to the speciality distribution of working doctors, and (2) what factors influence the young doctors' speciality choices. As shown in Figure 1, we hypothesised that gender and personality traits have an impact on speciality choice, and that career motivation as well as life goals have an influence, too. The present paper aimed to examine these hypotheses.

Residents' speciality choices at T ₃: Of the 522 physicians participating at the third assessment, 83 (15.9%) had not yet decided in which speciality they wanted to qualify. The remaining 439 residents had decided to qualify in the following specialities: 45 (8.6%) in primary care, 126 (24.1%) in internal medicine (including all sub-specialities of internal medicine), 68 (13.0%) in surgical disciplines, 31 (5.9%) in gynaecology & obstetrics, 40 (7.7%) in anaesthesiology and intensive care, 44 (8.4%) in paediatrics, 25 (4.8%) in psychiatry, and 60 (11.5%) in other specialities (such as dermatology, ENT, neurology, ophthalmology, radiology). The participants had a list of all 43 officially acknowledged medical specialities in Switzerland, marked with a code number, which they could fill in answering the question concerning the speciality choice.

The development of residents' speciality choices is shown in Figure 2. At the end of medical school 60% of the female and 51% of the male graduates had made their speciality choice. In their second year of residency, 71% of the female and 64% of the male residents had decided, and in the fourth year of residency, 83% of the female and 84% of the male participants named their speciality aspired to. The distribution of the different specialities within the gender group did not change much between the three measurements. However, the distribution of the specialities aspired to by the study participants is different from the speciality distribution represented by the working doctors who hold a speciality qualification. Compared to all working female specialists, there are significantly fewer female residents aspiring to become primary care physicians and psychiatrists; also the group of other specialities is smaller. It might be that some of the still undecided residents will choose one of those specialities later on. Looking at the male participants, fewer residents want to become primary care physicians, internists or psychiatrists but significantly more want to go into surgical disciplines compared to working male doctors.

The present study is part of an ongoing prospective survey of a cohort of Swiss medical school graduates. Subjects included in the study are fourth-year residents in different medical speciality fields. The aims of the study are to examine (1) the development of the residents' speciality choices since graduating from medical school and (2) what factors influenced their choices.

Development of the residents' speciality choices: at the end of medical school a considerable number of students have not yet developed precise ideas as to which speciality they want to work in. The main reason is lack of clinical experience. During the first and second years of post-graduate training they gain insight into various specialities, which makes it easier for them to make their decision. Some residents change the speciality they primarily aspired to, but not so many do. Nor is a marked shift away from one speciality towards another evident. Compared to the working doctors, significantly fewer young doctors aspire to become primary care physicians (PCPs) or psychiatrists. In the competition-based health care systems of Switzerland and the other German-speaking countries the professional prestige, social status, and income of other specialists is much higher than those of PCPs and psychiatrists. As long as the current health policy does not create better professional conditions, the shift away from PC and psychiatry will continue. If this trend prevails for some time, there will be a shortage of doctors providing basic somatic and mental health care, especially in rural areas.

Factors influencing speciality choice: As hypothesized, we found gender different speciality choices, female doctors being over-represented in specialities like gynaecology & obstetrics (G&O), paediatrics, and anaesthesiology and male doctors in surgical specialities. Similar results are reported in other studies [11–13, 17, 34]. Although G&O entails long hours and a heavy surgical workload, a growing number of women choose this speciality. Women are interested in surgical specialities, but often experience gender-relevant exclusion mechanisms in other surgical fields [17, 35]. The marked gender shift in G&O is due to the growing attitude, starting in the late 1980s, that women should be treated by female physicians [36]. Paediatrics, the other speciality mainly chosen by women, is also a speciality in which gender schemas play a certain role [14]. This gender-distinct speciality choice was already found when the participants were in the last year of medical school [15]. The reasons for an increasing number of female doctors choosing anaesthesiology might be manifold: anaesthesiology is a professionally prestigious speciality like surgery but not as competitive, it covers a broad medical spectrum and offers good options for part-time work and good promotion prospects, all factors appreciated by females. Whether the decisive role of gender is based on internalised gender roles or whether there are open or masked deterrents cannot be distinguished by this study.

Personality, career motivation, life goals and speciality choice: According to our assumption, we found that gender, personality, career motivation and life goals have an impact on speciality choice. Petrides and McManus [1] described a mapping of medical careers based on the typology found by Holland in careers in general. Holland's theory [37] suggests that careers can be organised into six broad types, which can be represented around a hexagon, known by the acronym RIASEC, standing for 'Realistic', 'Investigative', 'Artistic', 'Social', 'Enterprising' and 'Conventional'. They also referred to the 'Things↔People' and 'Ideas↔Data' dimensions proposed by Prediger [38] which can be underpinned to Holland's typology. Both models use the attribution of the medical specialities based on the characteristics of their professional activities. We suppose that residents choose a medical speciality in which they can conduct their professional activities corresponding to their special personality traits, career motivation, and life goals aspired to. The residents in surgical specialities in our study were characterised by high values for instrumentality, intrinsic and extrinsic career motivation, 'power' and 'achievement' as life goals. These attributes and attitudes are mainly 'Things/technique-oriented' according to Prediger and can be assigned to Holland's realistic career type. Participants choosing anaesthesiology and intensive care assessed themselves almost as highly instrumental as doctors in surgery. They can also be mapped to the realistic type. Contrary to the study by Petrides & McManus [1], the G&O residents in our sample revealed characteristics of 'People Orientation', matching the social type, a fact which might be influenced by the high number of females in this group. They stated high expressiveness and life goals aiming at satisfying social relationships. Paediatricians in our study, mainly females, showed similar characteristics as the G&Os. Psychiatrists differed from all the other specialities by rating the extraprofessional concerns highest, while expressiveness and altruism were not values very high. They could not easily be assigned to one of the RIASEC types. The same applies for primary care physicians. They showed characteristics of 'People Orientation' (high extraprofessional concerns and altruism), and rated in the medium range in most of the other aspects. The internists were allocated to the investigative type by Petrides & McManus [1], a type in which patient-relationships and diagnostic investigations play their part. This might also apply to the internists in our study; they described themselves as empathetic (adequate expressiveness), but also instrumentally-oriented. The RIASEC mapping of medical specialities did not consider career motivation and life goals but focused only on personality traits. Our results indicate that career motivation and life goals are even more important for the speciality choice than personality traits.

In summary, the results of our study indicate that gender plays a decisive role in speciality choice, while the influence of personality declines after controlling for career motivation and life goals. The feminisation of medicine and especially of some specialities can be expected to lead to fundamental changes in the medical system. One can assume that the style of leadership in hospitals will change: As long as men are department heads, a "command and control style of managing others" will predominate, while women tend towards "interactive leadership" [39]. Other aspects concern employment: more part-time jobs have to be provided for women doctors with family obligations. However, the feminisation also carries the risk of the danger of vertical and horizontal gender segregation [40]: female physicians often spend more time with patients while male doctors look at what is more advantageous for their career, such as laboratory work, developing research projects and writing papers. These differences in working attitudes result in gender-different career opportunities: males taking over leadership positions and females looking after their patients' needs.

The speciality choice of the new generation of doctors is a matter of concern for the health care system in Switzerland. Ideally, as many doctors should practise in a speciality as are needed to cover the population's health care. As the statistics of the Swiss Medical Association and our study data show, a lack of primary care physicians as well as of psychiatrists will occur in the next decade. The question arises whether incentives, mainly financial, will encourage more young doctors to choose to become primary care physicians or psychiatrists, or whether access to the speciality qualification programmes and to the allocation of private practice licenses should be regulated by health policy.

There should be an acknowledgment of the distinctive features of female physicians' careers. Workplace conditions should allow women doctors, if they want, to combine pursuing a prestigious career in medicine with having a family. There is a need to promote and encourage their instrumental traits and extrinsic career motivation. Motivational guidance by senior physicians throughout medical training should not only focus on the professional career but should also consider the personal life goals of those being mentored.