Change of Interest in Surgery during Medical School: A Comparison of Men and Women

Despite the increasing enrollment of women in medical school (45.6% for 1999-2000), women continue to be underrepresented in the field of surgery and surgical subspecialties, accounting for 7-20% of residents in recent years.¹ This trend is not confined to the United States, but is found in Canada, the United Kingdom, and The Netherlands as well.^2-5 Reasons for the disparity of surgery as a career choice between men and women are incompletely understood, but may involve lifestyle expectations and concerns about feasibility of raising a family, given the demands of the surgical profession.^2,4 Gender equity in the surgical professions will require a more complete understanding of the reasons women choose not to enter the field.

This study was designed to address the following questions: (1) Are there differences in the proportions of men and women who lose or gain interest in surgery during medical school? and (2) What are the variables that contribute to losing or gaining interest in surgery during medical school?

Method

The study sample included 4,676 physicians (3,386 men and 1,290 women) who graduated from Jefferson Medical College in the past three decades (1970-2000) and expressed their specialty preferences at the beginning of medical school in a matriculating questionnaire (71% of total graduates).

The study sample was classified into four groups. Group 1 consisted of those who planned to pursue a surgical specialty (defined as general surgery, orthopedic surgery, neurosurgery, otolaryngology, and urology) at the beginning of medical school, and entered one of the aforementioned surgical specialties (n = 357, 338 men, 19 women). Physicians in this group are those who retained their interest in surgery during medical school.

Group 2 was those who planned to pursue a surgical specialty at the beginning of medical school but entered a non-surgery residency (n = 401, 341 men, 60 women). Physicians in this group are those who lost their interest in surgery during medical school.

Group 3 was composed of those physicians who did not plan to pursue a surgical specialty at the beginning of medical school but who entered a surgery residency (n = 586, 508 men, 78 women). Physicians in this group are those who gained interest in surgery during medical school.

Group 4 consisted of those physicians who did not plan to pursue a surgery specialty at the beginning of medical school and entered a non-surgery residency (n = 3,332, 2,199 men, 1,133 women).

The dependent variables included undergraduate grade-point averages (GPAs) in science and nonscience courses, first- and second-year GPAs in medical school, grades in the third-year objective examinations in six core clerkships (family medicine, internal medicine, obstetrics-gynecology, pediatrics, psychiatry, surgery), ratings of clinical competence in these clerkships, score on medical licensing examinations such as Steps 1, 2, and 3 of the United States Medical Licensing Examinations (USMLE), formerly Parts 1, 2, and 3 of the National Board of Medical Examiners examinations, and professional peak income estimates at the beginning and at the end of medical school. The income estimates were adjusted for the inflation rate to 1999 dollar values by using the Consumer Price Index.

Also among the dependent variables were ratings of postgraduate clinical competence in areas of data-gathering and processing skills, interpersonal skills and attitudes, and socioeconomic aspects of patient care. These ratings were made at the completion of the first residency year by the residency program director or the person most familiar with the resident's performance.⁶ For the sake of simplicity in group comparisons, all the continuous dependent variables, with the exception of income expectations, were converted to a common standard distribution with a mean of 100 and a standard deviation of 10.

We used two-way analysis of variance (ANOVA, 4 × 2 design) in which the four aforementioned groups and gender (men and women) were the independent variables. Duncan's post hoc comparison test followed significant ANOVA results to identify significant pairwise mean differences.

Results

To test the representativeness of the study sample, graduates who specified their specialty areas of interest at the beginning of medical school were compared with the rest of their cohorts who did not specify their specialty choices, and thus were excluded from this study. Women and those who performed better in objective examinations of medical knowledge are overrepresented in the study sample.

To answer the first research question, we studied the cohorts of graduates who lost or gained interest in surgery during medical school. Of the 679 men who specified a plan at the beginning of medical school to pursue surgery, 338 (50%) retained their interest and entered a surgery residency. Therefore, the rate of losing interest in surgery among men was 50%. Of the 79 women who planned to pursue surgery at the beginning of medical school, only 19 (24%) entered a surgical residency. Therefore, the rate of losing interest in surgery among women was 76%. Interestingly, graduates of both genders with an initial interest in surgery who did not pursue surgery were not likely to pursue other surgical fields such as obstetrics-gynecology or ophthalmology. These graduates mostly pursued internal medicine (45% of men, 33% of women), family medicine (15% of men, 10% of women), and pediatrics (5% of men, 22% of women) in their first years of residency.

Conversely, we observed that the rate of gaining interest in surgery during medical school was 19% for men (508 of 2,707 men who did not plan to pursue surgery at the beginning of medical school entered a surgery residency), compared with 6% for women (78 of 1,211 women who did not plan to pursue surgery at the beginning of medical school entered a surgery residency).

These findings indicate that while the rate of losing interest in surgery during medical school for women is higher than that for men (76% for women, 50% for men), the rate of gaining interest in surgery during medical school for women is only about one third of that for men (6% for women, 19% for men). The association was statistically significant (χ₍₃₎ = 249.0, p < .001).

We examined the proportions of losing and gaining interest in surgery in two different time periods (1970-1990 and 1991-2000) to investigate whether any significant change had occurred. The proportion of women was significantly smaller in the 1970-1990 time period (19%) than in the 1991-2000 period (36%). No significant change in the general pattern of findings was found. Seventy-seven % of women and 54% of men lost interest in surgery in 1970-1990; 74 and 44% of women and men did so in 1991-2000. Proportions gaining interest in surgery during medical school also remained unchanged (6 and 17% of women and men in 1970-1990; 7 and 21% of women and men in 1991-2000).

To answer the second research question, we compared men and women in the four groups on a number of variables. Means of the variables for which significant differences were found are reported in Table 1. No statistically significant difference was observed on other dependent variables described previously.

Table 1

Main effects of grouping. Results of statistical analyses showed that, regardless of gender, those who maintained their interest in surgery during medical school (Group 1) and those who gained interest in surgery (Group 3) obtained significantly higher clinical competence ratings in the surgery clerkship.

There was no significant group difference in performances on Steps 1 and 2 of the USMLE; however, those who pursued surgery in their residency training (Groups 1 and 3) had significant lower average scores on Step 3 (or Part 3) than did those entering nonsurgical residencies.

We also found statistically significant differences in the professional incomes estimated at the beginning of medical school. The highest income was estimated by those who retained their interest in surgery (Group 1), followed by Groups 3, 2, and 4. The same pattern of findings was observed for the professional income estimated at the completion of medical school. It is interesting to note that those who did not retain their interest in surgery (Group 2) lowered their expectations of income from the beginning to the end of medical school, and those who gained interest in surgery (Group 3) increased their expectations of income. Women's income expectations were lower than men's income expectations for all four groups.

Main effects of gender. The ANOVA results showed that regardless of grouping, women obtained higher undergraduate GPAs in nonscience courses than did men. Women also had higher grades on the objective examinations in six core third-year clerkships, including surgery, on ratings of clinical competence given in the surgery clerkship, and on postgraduate ratings of competence in the area of socioeconomic aspects of patient care. Conversely, men scored higher than did women on Step 1 of the USMLE and estimated a significantly higher professional income than did women at the beginning of medical school, as well as at the completion of medical school.

Discussion

Our study finds that women at the beginning of medical school express significantly less interest than do men in pursuing surgery or a surgical specialty (6% of women versus 20% of men). Second, a higher proportion of women than men who plan to pursue surgery at the beginning of medical school lose their interest in surgery and pursue non-surgical specialties (76% of women versus 50% of men). Third, the proportion of women who gain interest in surgery during medical school is significantly smaller than that of men (6% of women versus 19% of men). This represents a triple barrier to correcting the current underrepresentation of women in surgical specialties.

Findings with regard to academic performances of women and men in medical school and postgraduate competence ratings, regardless of their specialty areas of interest, were consistent with our previous research.⁷ In addition, results of this study confirmed those in two other studies⁸ that found that USMLE Step 3 scores were significantly lower for physicians who pursued early specialization (e.g., surgery) than for their counterparts in broader medical specialty areas (e.g., family medicine or internal medicine).

Our study also shows that the rating of clinical competence in the third-year surgery clerkship is a significant predictor of gaining or losing interest in surgery, suggesting that enhancing or diminishing a sense of competence in surgical skills is important to students' career choice.

Women's expectations of lower professional income in general and higher income expectations among both men and women who planned to pursue surgery were in agreement with previous findings.⁹ The women's lower income expectations may reflect anticipation of reduced work hours to meet family demands. Many studies show that women, including women interested in surgery, are more concerned than men about the impact of child-rearing and family demands on their careers.^2,4,10

Our findings support other studies suggesting that women reject surgery as a career choice more often than do men.^4,11 and provide additional information that this occurs even among women who have an initial interest in surgery. Most studies cite lack of female role models and concerns about balancing career and family as main reasons for this trend.^2,4,10 Studies show that women who enter surgical careers are more likely to be unmarried and childless¹² and more likely to agree that personal life should be sacrificed for work.¹³ In addition, surgeons (both men and women) may have different personality traits than non-surgeons.⁵ Women who choose surgical careers, however, are as satisfied with their careers as are other women physicians.¹²

Limitations of this study are that the data were obtained from students graduating from one private medical school. The percentage of women who enter surgical fields from our medical school, however, does not differ from national statistics.¹ It is unlikely, therefore, that the culture of surgery at this school is significantly different from those at other medical schools.

Another important point that could change the outcomes is the classification of surgery in this study. One may argue that it could be desirable to study only those who were interested in general surgery by excluding other surgical specialties and subspecialties to remove any confounding effects. This approach was not practical in this study because of the smaller number of women in Group 1. Also, redefining surgical specialties by including obstetrics-gynecology could be another approach in studying the outcomes.

In conclusion, efforts to increase the numbers of women in surgical careers must identify perceived and actual barriers to successful careers in surgery. In addition, factors contributing to change of interest should be determined. Recruitment efforts should target women before and during medical school, when specialty choice is most often determined. Women who express an initial interest in surgery are still nearly four times as likely to enter surgery as are women without an initial interest. For medical schools and departments of surgery wishing to enhance interest in surgical careers among women, focusing efforts on this group may be worthwhile. An important follow-up of this study is planned to ascertain the subjective experiences of students and factors that prompt women to gain or lose interest in surgery.