Medicine used to be a predominantly male occupation, but an increasing influx of females into medical schools is changing the future demographics of physicians [1, 2]. Today, females account for half of all medical students in the USA [2] and are in the majority in several European countries [3]. This demographic shift influences specialty preferences. Females tend to be less attracted to surgical specialties [1, 2, 4], which could lead to staff shortages among surgical specialties [4].

Genetic, hormonal, and environmental differences between males and females manifest in a complex manner. For example, although anatomical studies suggest gender differences in the brain [5, 6], the functional value of these differences is unclear [6–10]. Psychomotor studies have found gender differences in solving practical tasks [11]. Visuospatial gender differences exist [12, 13], and there is even evidence to suggest a gender-specific difference in the interaction between psychomotor learning and hormonal response [14, 15]. Understanding the different ways in which people learn is crucial in order to better design the surgical education of future that accommodates the educational needs of both genders.

The aim of this study was to systematically review studies that investigate gender differences in the acquisition of surgical skills among medical professionals.

Materials and methods

We followed the PRISMA guidelines for reporting systematic reviews [16].

Eligibility criteria

We included studies that investigated the role of gender in the acquisition of surgical skills. All types of published journal articles were included (randomized controlled trials, non-randomized comparative studies, case–control studies, cohort studies, multicenter studies, descriptive studies, evaluation studies, retrospective studies, clinical trials, and other journal articles). We excluded studies that only stated the presence or lack of gender differences without any further specifications. We excluded abstracts with no articles and studies in languages other than English and Danish.

Search strategy and study selection

We searched the PubMed database using relevant search terms and Medical Search Headings and Subheadings (MeSH) combined with Boolean operators (Table 1). We compiled the search on the April 26, 2014. Two authors (AA and LK) independently performed the searches and selected relevant articles that met the eligibility criteria. Differences were solved through discussion until a consensus was achieved. The reference lists of all included studies were examined for additional studies that met our eligibility criteria.

Table 1 Search terms and the number of studies identified
Full size table

Data collection and synthesis of results

We extracted data on the year of publication, country of origin, study design, study population characteristics including employment and gender distribution, means of testing, and study results. Due to the heterogeneity of the study questions, we were unable to perform any meaningful meta-analysis. All authors (AA, YS, CR, and LK) contributed to the qualitative synthesis of the results.

Results

Our electronic search identified 247 publications (Table 1; Fig. 1), of which 14 remained after removing duplicates and obviously irrelevant articles. All 14 publications were reviewed in full-text format. Using the reference lists, we identified five additional publications of interest. From the total 19 publications that were reviewed in full text, one was excluded due to our eligibility criteria. A total of 18 studies were included in our qualitative synthesis with a total of 2,106 study participants (Table 2).

Fig. 1
figure 1

Flow chart of the study selection process

Full size image
Table 2 Characteristics of included studies
Full size table

The included studies were comprised of nine cohort studies (eight experimental and one observational), five cross-sectional studies (three experimental and two observational), two randomized controlled studies, one randomized controlled cross-over study and one non-randomized controlled study. The gender distribution ranged from 11 to 61 % females, and one study did not report the gender distribution. Nine studies were from the USA, two from Canada, two from the UK, two from Denmark, two from Sweden, and one from Switzerland. Fourteen studies used simulated surgery, two were based on self-reported information on surgery, one relied on Medical College Admission Test (MCAT) results and psychometric tests, and one study used an isometric strength testing unit. The authors agreed to present the studies using five categories based on the study population in order to systematize the presentation of the studies and present the tendencies that generally correlated within the chosen study subjects:

  • Medical students [17–24].

  • Both medical students and residents [25–27].

  • Residents only [28–30].

  • Gender differences in needed physical strength [31, 32].

  • Other gender-related training conditions [33, 34].

Studies on medical students

Studies on medical students were based on simulated laparoscopy and virtual reality (VR) simulators (Table 3). Two of the studies focused on gender-specific anatomical and psychological predispositions [17, 18]. Visuospatial ability correlated with performance on a VR simulator among right-handed males, but not among females [18]. In another study with a laparoscopic simulation setting, males showed a higher degree of ambidexterity and higher precision using their dominant hand; this is suggested to be due to the standard sizes of the instruments, which may not have been designed for smaller hands among females [17]. Other studies have looked into the students’ fields of interest. For example, males generally perform better on VR simulators [18–20], but scores also correlate with gaming experience, regardless of gender [18, 19]. Simulator performance is also correlated with students’ preference of surgery as their future specialty, regardless of gender [20, 21]. Even when adjusting for factors such as gaming and preference for surgery, which may differ demographically between the genders, males seems to have a significant advantage [18, 19, 21].

Table 3 Studies on medical students
Full size table

Other studies focused on the skills learning environment [22–24]. The performances of female medical students correlated with their theoretical knowledge, which is not the case for males [22]. When learning suturing, males showed a preference for repeated practice, while females preferred one-on-one instruction [23]. Instructor feedback versus no instructor feedback in VR simulator training demonstrated an interesting gender-specific difference in impact: without instructor feedback, males outperform females, whereas females and males performed equally with instructor feedback [24]—this indicates that females respond better to instructor feedback [24].

In summary, male medical students outperform female medical students, partly due to gender-specific interests, such as gaming and preference for surgery. Studies also suggest that observed differences could be due to gender-related differences in learning preferences.

Studies on both medical students and residents

Three studies included both medical students and residents (Table 4). On VR simulators, males tend to outperform females [25, 26]. Interestingly, gender differences cease to exist if gaming is included as a factor in subgroup analyses [26]. Another study compared initial and final completion time of Fundamentals of Laparoscopic Surgery tasks [27]. Initially, males received higher performance scores, but females responded better than males to training and gender differences eventually disappeared [27]. Thus, consistent with previously mentioned studies on medical students, studies on both medical students and residents have also found an effect of gaming experience and learning environment.

Table 4 Studies on both medical students and residents
Full size table

Studies on residents only

Three studies included residents only (Table 5). One study found that female residents outperformed male residents in academic achievements measured on MCAT, National Board Examinations, and verbal tests. Males, on the other hand, achieved higher scores on visuoperceptive skills and on confidence and task organization [28]. On a VR-simulated laparoscopy, male residents completed tasks faster than females, but the number of errors and unnecessary movements were similar [29]. On a box training laparoscopy simulator, no gender differences were found, and after a 4-week skill training program, one study was initially unable to find any gender differences [30]. After a year, however, retention was significantly higher among females [30]. Results at resident level are more mixed and do not show any clear gender differences.

Table 5 Studies on residents only
Full size table

Studies on gender differences in required physical strength

Two studies investigated the amount of strength required to perform procedures (Table 6). A study on obstetric forceps delivery showed that although male residents are able to generate more force than female residents, the differences are above any clinically meaningful level and are therefore insignificant for clinical practice [31]. Similarly, a study on intubation found that females do not lack the necessary physical strength and found no significant gender differences [32].

Table 6 Studies on gender differences in needed physical strength
Full size table

Studies on other gender-related training conditions

Two studies explored gender-related behavior (Table 7). A study on obstetrics residency review forms in the USA found that male residents had higher forceps delivery rates [33]. Another study investigated the influence of demographic factors on cataract surgery training and found that females were less likely to access surgical training facilities [34]. After surgical training, females operated fewer patients per week and had a lower overall number of surgeries [34].

Table 7 Studies on other gender-related training conditions
Full size table

Discussion

The findings of our systematic review suggest the need to realize that males and females are not alike when it comes to acquisition of surgical skills. Male and female medical students differ in their visuospatial abilities, interest in surgery, and gaming experience [18–21]. Providing females with instructor feedback and one-to-one training seem to eliminate these differences [23, 24, 27, 30]. We also found that there are no meaningful differences in the level of strength required for surgery [31, 32], but there do seem to be gender-specific differences in surgical work practice [33, 34].

Male medical students are initially more confident about their surgical abilities and take more risks [18, 28, 35, 36]. Female medical students tend to have longer reaction time with higher precision [11, 23, 37, 38], which can be described as risk-averse behavior. This may explain why males perform better initially. This perception of males being risk-takers and females being risk-averse seems consistent within a wide range of behavioral studies, ranging from motor vehicle collisions to financial investments [39].

In the acquisition of surgical skills, male medical students performed better and studies found the greatest gender differences in the visuospatial abilities and speed [20, 22, 23]. Similar observations on visuospatial differences were also found in other psychomotor studies outside of the medical profession [12, 13] and suggest a brain dimorphism on a general population level. There may also be a cultural influence, as males may be more likely to have played video and ball games during their childhood, which could help develop their visuospatial abilities. Cross-cultural studies on this matter are warranted.

Studies on residents showed mixed results, and female residents even outperformed males in retention tests [29, 30]. This may be due to a selection process into surgical specialties. Physicians accepted into resident programs may be more homogeneous in areas such as interest and gaming experience, which may explain the less clear differences in resident-only studies.

Female residents and medical students acquire surgical skills markedly better when trained appropriately [23, 27, 30], which includes one-on-one training with clear instructor feedback [23, 24]. Therefore, a key message from this systematic review is that surgical training of females should include more supervision.

Female physicians are less likely to perform surgical interventions in clinical practice than males [33, 34]. We have no reason to believe that this is due to a lack of strength [31, 32]. Instead, studies suggest that female physicians may be discriminated against during surgical rotations [40]. Firstly, this is unacceptable for purely ethical reasons. Secondly, considering the higher rate of female medical students [4], this will naturally lead to staff shortages in the surgical specialties unless such discrimination is addressed. Mentoring initiatives are currently being implemented as a way of attracting a higher number of female colleagues [41].

A systematic review is only as good as the included studies, and the consequent limitations of this study should be noted. The number of studies is low, most of the studies have small sample sizes, and the diversity of their design does not permit a quantitative comparison of outcomes. Previous life experience may have an influence on surgical abilities; for example, several studies have highlighted the value of gaming experience. We suggest that future studies should investigate the influence of other experiences that have not been highlighted. All but one of the studies is based on simulated surgery, which does not represent the true diversity of the surgical world. Even though simulators provide a great tool for repeatable comparison, the abundant number of variables that prevails in reality is overlooked, which leads to an oversimplified amount of evidence to explain what is a complex topic. Finally, all studies were conducted in Western countries, which may not represent the realities in the rest of the world.

In conclusion, male medical students seem to outperform female medical students, which may be due to differences in their visuospatial abilities and their varying interests that are associated with better acquisition of surgical skills. This initial difference disappears early in a surgical career, probably due to a selection bias. While males are more willing to practice on their own and take the associated risks, females prefer mentorship and one-on-one instructor feedback. Most importantly, surgical training eliminated initial gender differences. Therefore, we recommend personalized surgical training programs that acknowledge the different needs of participants, such as mentorship and one-on-one training. Gender issues should be considered when designing surgical training and curricula to better accommodate the needs of future surgeons.