The influence of early clinical experiences on career preference of male and female medical students

Abstract

Background: Clinical experience is considered to affect medical students’ career preferences. It is not known whether the sequence of the clinical rotations influences these preferences.

Aim: To explore whether the first clinical clerkship has more impact on career preference than the second by examining the association between the first clinical clerkship and the choice of an elective sixth-year internship.

Method: University Medical Center Utrecht students are assigned to either a surgical or a medical ward for the first third-year clerkship and to the other ward for the second clerkship. In a retrospective cohort study, internship data of 488 sixth-year students were related to their first clerkship 3 years earlier.

Results: For the group as a whole, no association was found between third-year clerkship and sixth-year internship. However, male students who had been assigned to surgery first more often chose a surgical internship than those who had been assigned to medical clerkship first and vice versa (p < 0.02). Within the female subgroup, no association was found.

Conclusion: A positive association between the nature of the clerkship and the sixth-year internship preference among male students suggest that the first clinical experience can affect later specialty preference.

Introduction

A well-functioning health system benefits from correspondence between required medical personnel in the system and medical students’ preferences for specialties. In reality, however, certain fields face a relative shortage of applicants, while vacancies in other specialties are over-subscribed (Pronk 2007). It is known that some clinical specialties are more popular among students than others (Wright et al. 2004).

Factors believed to generate interest in a certain medical specialty are manifold (Azizzadeh et al. 2003; Barshes et al. 2004; Lambert & Holmboe 2005). A factor considered particularly important is clinical experience (Morrison & Murray 1996; Williams et al. 1997; Soethout et al. 2007). The length and quality of this experience, including the nature of the department offering the rotation, could add to the fostering of interest for a particular specialty (Lambert & Holmboe 2005).

It is not known whether the position of a rotation within the curriculum could be a factor of importance too. One could argue that the first experiences with patient care could be so overwhelming that it could have a disproportionate impact on subsequent career interest (Paiva et al. 1982).

This possible stronger effect could be either positive or negative; a clear hypothesis about its direction was not found in the relevant literature (Arora et al. 2006). Potts & Brazeau (1989) examined the effect of the first clinical rotation on specialty choice and postulated that the likelihood of a student choosing a career matching the specialty of the initial clerkship would be small. They assumed that the students’ new role as a clinical clerk would make the student feel relatively uncomfortable with the specialty encountered first. The authors could not substantiate this assumption, as they found the sequence of clinical clerkships of little importance in final decisions regarding residency selection.

On the other hand, a more recent review by Barshes et al. (2004) showed that the order of clerkships was relevant indeed, but particularly for students who had not yet decided on their specialty choice: 73% of medical students undecided on specialty choice and 80% of students who change their choice of specialty during the third year of clerkships eventually chose specialties they had experienced during the first half of the third year.

The topic of the present study is the possible influence of the order in which the first clinical clerkships are followed on the preference for a specialty by the end of curriculum. We postulate that the first clerkship presented in a series of clerkships, will be more influential on future career preference than the clerkships presented later in the series. We believe such a positive association is justifiable, since medical students are usually eager and enthusiastic to gain clinical experience after 2 years of mere theoretical training.

One characteristic of the Utrecht curriculum is the early scheduling of two 6-week clinical clerkships in the third year of a 6-year course: Internal Medicine and General Surgery (Kamalski et al. 2007). The students are assigned to do these third year clerkships in a random order. At the end of their fifth course year, Utrecht students make a personal choice for a 12-week internship during their sixth year. Most students choose to gain experience in a potential field of later specialty. The sixth-year internship often is the first step towards a specialist career, and is not as heavily limited by capacity restrictions as is the case with vacancies of resident training positions. Therefore, we considered the choice for the sixth-year internship a better indication of specialty preference than the actual career destination.

This study presents a quantitative analysis regarding the possibility that students scheduled to have Internal Medicine as their very first clerkship, will later prefer a medical career, whereas students scheduled to have General Surgery as their very first clerkship would be more inclined to pursue a surgical career. The presence of such an association might allow for curricular adjustments to influence students’ medical career choice towards the needs of the healthcare system.

Table

Practice points

The first clinical experience seems to affect internship preference of male but not of female senior medical students. For female senior medical students, lifestyle controllability may be a more important factor than the first clinical experience for internship preference.

Method

Third-year medical students were assigned by the department of Student Affairs to either General Surgery or Internal Medicine first for a six-week clerkship. Students’ preferences were not taken into account in this procedure. Three years later, at the end of their fifth year, students registered for a 12-week internship of their preference.

The data used for this study were collected retrospectively from the faculty's administrative database in April 2007. Data regarding the content of the first clinical clerkship and the preferred sixth-year internship were collected of 488 sixth-year medical students at University of Utrecht from the cohorts of 2004/2005, 2005/2006, and 2006/2007.

The sixth-year internship preferences were categorized as either “surgical” or “medical.” The specialties classified as surgical internships included general, plastic and cardiothoracic surgery, gynecology and obstetrics, neurosurgery, orthopedics, and urology. The specialties classified as medical internships were general internal medicine, gastro-enterology, rheumatology, hemato-oncology, pulmonary disease and tuberculosis, cardiology, geriatrics, infectiology, intensive care, pediatrics, (pediatric) rehabilitation, neurology, and psychiatry. Logistic regression, chi-square (χ²), and Mann–Whitney-U (MWU) analyses performed with SPSS 15.0 were appropriate, with statistical significance defined as p < 0.05. Results (percentages) are presented for sixth-year internships with either a medical or surgical nature only, considering these results are the inverse of one another.

Results

General characteristics

Of the total study population of 488 students, two students were excluded because of missing data regarding their third-year clerkships and one student because of missing data regarding the preferred sixth-year internship. The remaining 485 subjects consisted of 156 male (32%) and 329 female students (78%) with an overall median age of 25.4 years (23.1–57.4; N = 485) when they started their graduation year. Male students were somewhat older (median 26.1; range 23.6–38.2) than their female counterparts (median 25.3; range 23.1–57.4; MWU p = 0.000) when they registered for their sixth-year internship. At the start of the first clerkship during the third year, this age difference was not present (median age 21.5 years; range 19.5–42.4; N = 441), suggesting that it took male students longer to finish the third, fourth, and fifth year. A total of 38 students were aged 30 or above during the sixth year. However, the gender distribution did not differ significantly between students below and above this age (respectively 32 and 40% male; χ² p = 0.315).

Third-year clerkship assignment and choice of sixth-year internship

For the third-year clerkship, 370 students (76%) were assigned to follow Internal Medicine first, against 115 students (24%) who were assigned to follow General Surgery first (similar distribution of gender and age over groups; both p > 0.05). For a sixth-year internship, 339 students (70%) opted for a medical specialty as first choice whereas 146 students (30%) chose an internship with a surgical nature. Students who chose an internship with a surgical nature were, in general, older (median 26.0; range 23.6–38.2) than their counterparts who chose a medical specialty (median 25.3; range 23.1–57.4; MWU p = 0.001). This difference is associated with the fact that male students were older and more often chose sixth-year internships with a surgical nature (N = 67; 43%) than their female counterparts (N = 79; 24%; χ² p = 0.000). Within the male population, students who chose a surgical specialty were not significantly older (N = 67; median 26.3; range 23.8–38.2) than those who preferred a medical specialty for their sixth-year internship (N = 89; median 26.0; range 23.6–36.8; MWU p = 0.404).

Of all students having done their third-year Internal Medicine clerkship first, 264 (71.4%) preferred a medical sixth-year internship. Of the students whose first third-year clerkship was General Surgery, 75 (65.2%) preferred a sixth-year internship with a medical character. These percentages do not differ significantly (Table 1). However, the male students who were assigned to do General Surgery first during their third year (N = 41) more often chose a sixth-year internship with a surgical character (N = 24; 59%) than their male counterparts, who were assigned to Internal Medicine first (N = 43/115; 37%; χ² p = 0.019). Consequently, the inverse was true for preferring a medical sixth-year internship when male students were assigned to Internal Medicine first during their third year. However, this association was not observed in the female student population (Surgical internship: N = 16/74 (22%) and N = 63/255 (25%); χ²p = 0.584) (Table 1).

Table 1.  Association between initial clerkship assignment and general choice of internship in sixth year

Initial third year clerkship assignment
	Internal medicine N = 370 (%)	General surgery N = 115 (%)	χ^2	p-Value
Internship with medical nature (sixth year)	264 (71)	75 (65)	1.57	0.210
Internship with surgical nature (sixth year)	106 (29)	40 (35)
Male students	N = 115	N = 41
Internship with medical nature (sixth year)	72 (63)	17 (41)	5.515	0.019
Internship with surgical nature (sixth year)	43 (37)	24 (59)
Female students	N = 255	N = 74
Internship with medical nature (sixth year)	192 (75)	58 (78)	0.299	0.584
Internship with surgical nature (sixth year)	63 (25)	16 (22)

Surgical internships included general, plastic and cardiothoracic surgery, gynecology and obstetrics, neurosurgery, orthopedics, and urology. Medical internships were general internal medicine, gastro-enterology, rheumatology, hemato-oncology, pulmonary disease and tuberculosis, cardiology, geriatrics, infectiology, intensive care, pediatrics, (pediatric) rehabilitation, neurology, and psychiatry.

A possible interaction between gender and the first clerkship during the third year was tested for significance using logistic regression analysis (block entry) with females and Internal Medicine as reference group for gender and first clerkship (indicator contrast) and sixth-year internship preference coded as surgical (0) or medical (1). Inclusion of gender alone at the first step was highly significant (p < 0.001). At step two, addition of the first clerkship variable did not contribute significantly (p = 0.275). However, addition of the interaction of gender and clerkship at step three was significant (OR 0.356; 95%CI 0.137–0.926). Table 2 presents the results of the regression analysis at step three.

Table 2.  Logistic regression analysis of the interaction between gender and initial clerkship assignment in predicting sixth year internship preference

	OR	95% confidence interval	p-Value
Gender	0.55	0.34–0.88	0.013
First clerkship	1.19	0.64–2.22	0.585
Gender * First clerkship	0.36	0.14–0.93	0.034

Females and Internal Medicine served as reference groups for gender and first clerkship (indicator contrast) and sixth-year internship preference coded as surgical (0) or medical (1).

Discussion

In the present study, we investigated whether 485 medical graduate students’ career preferences were influenced by their first clinical experience (medical or surgical) 3 years earlier. The read out for career preference was the preferred sixth-year internship, which is often followed by a resident position after graduation.

Our results first showed that the nature (surgical or medical) of the preferred sixth-year internship of 159 male students was positively associated with the field of specialty of their first clinical clerkship 3 years earlier (General Surgery or Internal Medicine). Interestingly, in 329 female students no such association could be detected. Secondly, we found that male students were 9.5 months older than female students at the start of their graduation year whereas this difference was not present at baseline. We will speculate on an explanation for these findings.

First, male students may be affected differently by their first clinical experience than female students, because of internal (personality, ability, family background, expectations, ideas, values and level of personal development, etc.) and external factors (a lack of career guidance mentorship, role models, gender discrimination, harassment, etc.) (Kruijthof et al. 1992; Hays 1993; Neittaanmaki et al. 1993; Field & Lennox 1996; Rademakers et al. 2008; Soethout et al. 2007). For instance, Field & Lennox (1996) studied medical students’ perceptions on the effect of their gender on their career choice. Overall, male students felt that their gender had no effect on their career choice, whereas the majority of female students thought it did. Some of the female students attributed this to the experience of being actively discriminated against during their studies.

A second possible explanation for our findings is that, when deciding career preference, female students may have other priorities and values than male students. Harris et al. (2005) conducted a survey among 4259 Australian medical graduates (average age 32.5) and found that factors of particular importance to women when choosing a field of specialty, compared with men, were appraisal of domestic circumstances, working hours and the opportunity to work flexible hours. A survey conducted by Soethout et al. (2007) among 1091 Dutch doctors who graduated between July 1999 and June 2002 revealed similar results. These observations are supported by findings of a survey conducted between 1990 and 2003 among all American medical graduates, described by Lambert & Holmboe (2005). In this survey, the interest in all so-called controllable lifestyle specialties (Schwartz et al. 1990), such as anesthesiology, dermatology, emergency medicine, neurology, ophthalmology, otolaryngology, psychiatry, and radiology increased among female medical graduates between 1996 and 2003. However, the interest of male students in controllable lifestyle specialties increased only in three of eight specialties between 1996 and 2003. Furthermore, in 2003, there was a slight convergence of the proportions of men and women choosing the controllable lifestyle specialties, which is perhaps a first indication of similar preferences (Lambert & Holmboe 2005). It is possible that the female medical students in our study more often considered it important that their sixth-year internship should be a controllable lifestyle specialty, which overruled or diminished the observed effect of the first clinical experience during their third year.

Our study has some limitations. First, the outcome in this study (career preference as indicated by the choice for the sixth-year internship), is only a midpoint in the path to ultimate completion of specialist training and providing specialist services. However, given shortage of positions in one field and abundance in another, we think that in our study career preference is a more precise measure of the motivations of medical graduate students. Secondly, timing is only one aspect of the way clinical experience influences career preference. We were not able to adjust our findings to other factors that influence career choice in medical or surgical specialties, like the exact content of the clinical experience, student satisfaction and preference for (un)controllable lifestyle specialties. Additional prospective, qualitative studies are needed to determine the importance of the first clinical experience.

In conclusion, our results show a positive association between the first third-year clerkship (Internal Medicine or General Surgery) and the nature (surgical or medical) of the sixth-year internship preference of male students. In female students no such association could be detected. These results suggest that the nature of the first clinical experience influences male and female students differently. Still, this could offer universities the ability to influence career preference among male medical students. Possibly in the future, this effect will be similar for female students, when all specialties will be (more) equally attractive in lifestyle controllability.

Acknowledgements

Our thanks to E.C. Gies, M. Koch, and A.J. Dikker from the faculty administration for supplying the relevant data, the reviewers of Medical Teacher for their helpful comments, and Dr Paul Oosterveld for methodological advice.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Additional information

Notes on contributors

Luc. E. Coffeng

LUC E. COFFENG, MD, is a PhD student at the Department of Public Health of Erasmus Medical Center, Rotterdam, and at the time of this study was a graduate medical student at the University Medical Centre Utrecht.

Anouck J.E. Visscher

ANOUCK J.E. VISSCHER, MD, is currently at Jellinek Mentrum, ACT Rehabteam, Amsterdam, and at the time of this study was a graduate medical student at the University Medical Centre Utrecht.

Olle Th.J. Ten Cate

OLLE TH.J. TEN CATE PhD, is a professor of medical education and director of the Center for Research & Development of Education at the University Medical Center Utrecht.