Factors influencing the choice of specialty of Australian medical graduates

For the analysis, 17 specialties were grouped into 11 categories (Box 1). Four specialties (public health medicine, rehabilitation medicine, occupational medicine and medical administration) with small numbers and closely aligned with adult medicine were merged with adult medicine. Intensive care respondents were merged with anaesthesia, and dermatology was excluded because of small numbers (adjusted baseline number of doctors, 7851).

Intrinsic characteristics included age, sex, marital status, dependent children, and location of medical school (Australia or overseas). Extrinsic factors included hours of work and full-time/part-time training program status. These were examined because they were considered to be an indication of the working conditions and availability of flexible training arrangements.5,6

The questionnaire was tested in a pilot study involving 66 participants from three vocational training programs.

Specialties with the largest representation were adult medicine, general practice, surgery and anaesthesia (Box 1) and these profiles were similar to those of the target population.18 Respondents’ average age was 32.5 years, 31.5% had dependent children, almost 70% had partners, and 15% graduated from an overseas medical school. Not surprisingly, across specialty groups there was a significant variation in the demographic characteristics of respondents (P < 0.01 for all variables). Of the 625 respondents who graduated from an overseas medical school, 603 (96%) were born overseas.

In total, 9.7% of respondents were training on a part-time or job-share basis, and across training programs there was wide variation in the proportion of doctors training part-time (P < 0.01) and in the number of hours worked on average per week by full-time trainees (Box 2).

Respondents were asked to indicate at what stage they chose their current specialty (eg, before entry to medical school, by the end of medical school, or the postgraduate year when this decision was made). In total, 37% of doctors had made their choice by the end of postgraduate year 1 (PGY 1), 43% during PGY 2 or 3, while a further 9% made this decision during PGY 4 or 5. For 11% of doctors, choice of specialty did not occur until 6 or more years after completion of their basic medical degree (Box 3). Trainee surgeons tended to choose their specialty earlier than other trainees. No differences were observed between male and female trainees in the timing of the decision about a specialty training program, while more recent graduates (eg, those graduating in 1996–2000) tended to decide on their specialty earlier in their career than graduates before this time (P < 0.01).

Respondents rated on a scale of 0–4 their estimate of the influence of each of the factors listed in Box 4 on their current choice of specialty (0 = not at all; 4 = a great deal). Three intrinsic factors were rated as having the greatest influence: “appraisal of own skills and aptitudes”, “the intellectual content of the specialty”, and “interest in helping people”. These were followed by seven extrinsic factors, the most important of which were the “atmosphere/work culture typical of the specialty”, “work experience since graduation” and “opportunity to work flexible hours”.

Differences across training programs in the weighting ascribed to the 23 factors influencing choice of training specialty, controlling for sex, age, marital status and other background characteristics, are shown in Box 4 and Box 5. The first of two sets of analyses explored differences between general practice trainees and trainees in the other specialties (as a group), while the second examined differences among specialties, excluding general practice trainees. These analyses indicated that general practice trainees, compared with other trainees, ascribed greater importance to “interest in helping people”, “appraisal of own domestic circumstances”, “opportunity to work flexible hours”, “number of years required to complete training”, and “experience of specialty as a medical student”.

Trainees in emergency medicine, anaesthesia, psychiatry, pathology and ophthalmology were more likely to ascribe importance to “opportunity to work flexible hours”, while this factor was of less importance to trainees in obstetrics and gynaecology and adult medicine. Trainees in surgery ascribed greater importance to the influence of consultants/mentors than trainees in the other nine training programs. Not surprisingly, “opportunity to do procedural work” was of greater importance to trainees in surgery, obstetrics and gynaecology, ophthalmology, anaesthesia and emergency medicine than to trainees in the other programs.

Differences were observed between male and female doctors in the weighting ascribed to the various extrinsic factors influencing choice (P < 0.01). Factors of particular importance to women compared with men were “opportunity to work flexible hours” (67% [1307/1951] of women rated this factor as important compared with 45% [1003/2253] of men; OR, 2.6), and “hours of work” (61% [1193/1952] of women, 47% [1066/2255] of men; OR, 1.8). Differences in rankings were also observed based on age, marital status and dependent children (or not). Of greater importance to younger trainees (< 30 years v ≥ 30 years) was “experience as a medical student” (respectively, 50% [678/1363] and 35% [1018/2939]; OR, 1.6), while factors of more importance to partnered doctors than to single doctors were “hours of work” and “opportunity to work flexible hours” (respectively, 55% [1607/2911] and 50% [602/1198]; OR, 1.3; and 57% [1650/2910] and 51% [609/1195]; OR, 1.3). Not surprisingly, “appraisal of own domestic circumstances” was of greater importance to doctors with children than to doctors without children (respectively, 54% [701/1306] and 38% [1103/2873]; OR, 1.7).

The strengths of our study include an acceptable response rate, with just over half of the target group responding, and these respondents shared the most salient characteristics of the target population (ie, completed medical training, registered with an Australian clinical college training program, and a similar specialty and sex profile).18 The validity of the questionnaire was strengthened through the use of focus groups, and it was pilot tested to inform and test its contents and design. The study’s weaknesses include that the reasons for non-response are not known, plus problems known to be associated with the use of structured questionnaires.19

The findings of our study are strengthened by the fact that they are similar to those of Davidson et al who reported that career choices of UK medical graduates were influenced by appraisal of their own skills and aptitudes and enthusiasm for the specialty.16 As in our study, these researchers noted that women doctors were more influenced by domestic circumstances and hours of work and working conditions than were male doctors. Our study extends knowledge about factors influencing choice of specialty by paying particular attention to factors amenable to change and to variation in these factors across disciplines.

Our findings suggest that, for training programs and teaching facilities with recruitment problems, there should be a review of the work culture typical of the specialty, the opportunity for trainees to work flexible hours, the quality of supervision/mentoring provided to trainees, and trainee satisfaction with the hours they are required to work. Given the importance that women ascribe to being able to work flexible and reasonable hours, it is hardly surprising that several specialties are presently attracting a comparatively greater proportion of women into their training programs. These specialties have training programs that provide real opportunities for part-time training and, on average, trainees work about 45 hours per week. Our findings also suggest that these factors gain in importance as doctors grow older and have partners. Hence, these factors may become increasingly important to male doctors in the future, given the increasing age at which students now commence their medical studies.8,11

Our findings confirm that observations during medical school and PGY 1 are of great importance in medical graduate decisions about choice of specialty. Variation across specialties in the timing of specialty decision suggests that some disciplines, such as surgery and, to a lesser extent, general practice, ophthalmology and paediatrics and child health, may have developed more effective strategies during the medical school and PGY 1 years than other disciplines in assisting graduates to select their specialty.