Teaching the physical examination to medical students is an important and integral part of their medical education. The pelvic, breast, and male genital examinations are especially challenging to teach and learn. The sensitive nature of the exam requires special interpersonal and technical skills on the part of teachers and learners as well as the consent and cooperation of the patient.
There is no consensus regarding the minimal number of examinations required to be confident and competent in performing the pelvic, breast, and male genital exams. However, for any given preclinical, didactic experience (video tapes, models, lectures, etc.), it might be assumed that the sheer number of examinations from these experiences correlates with confidence. During the clinic years, the type and degree of supervision may be important, in addition to the number of patients examined. Some instructors may repeat the students' examinations; others may observe them without repeating and confirming findings. We are not aware of any studies addressing what role the type of supervision plays in students' experiences in this area.
The student's own gender may also affect opportunities for learning skills for gender-based exams. Some studies have found that the gender of both the preceptor and the student influence the student's clinical experiences in gender-specific exam skills.1–5 Emmons and colleagues6 found that 68% of male medical students in a third-year obstetrics–gynecology clerkship reported gender discrimination. They performed fewer procedures compared to students who did not report gender discrimination, but the differences were small and they still achieved their clerkship requirements.
We designed our study to survey medical students' self-reported experiences in performing the pelvic, breast, and male genital examinations. We sought to evaluate the role of student gender, number of examinations performed, and patterns of supervision, on self-rated confidence in performing gender-specific examinations. We hypothesized that gender influences medical students' experiences in performing gender-specific examinations and that students with more experience and supervision would be more confident in performing these examinations.
In October 2001, we mailed a questionnaire to all 402 third- and fourth-year medical students from the University of Washington School of Medicine graduating classes of 2002 and 2003. In the questionnaire, they were asked to approximate the number of pelvic, breast, and male genital examinations they had performed by checking one of the following four categories: <5, 6–10, 11–15, and > 15. They were asked to estimate the percentage of these examinations that had been observed by a supervising physician and the percentage of exams that had been repeated by a supervising physician by checking one of the following four categories: <25%, 25–50%, 50–75%, and >75% for each question. We did not provide definitions of the examinations. Students were asked to select if they were either confident or not confident (binary response) in performing each of these examinations. We did not use clinical teachers or other forms of documentation to validate the responses. Students were also asked for their gender.
We entered the questionnaire results into a database and analyzed them using SPSS Version 11 (SPSS, Inc., Chicago, IL). We performed a chi-square analysis (4 × 2 chi-square tables) by gender and level of training for each type of examination regarding the number of examinations performed, percentage of examinations observed by a supervising physician, and percentage of examinations repeated by a supervising physician. A chi-square analysis (2 × 2 chi-square tables) was performed for each type of examination for the questions assessing confidence. We did a stepwise multiple regression analysis to determine predictors of confidence in performing each of the examinations using the following variables: gender, number of examinations performed, number of examinations observed by a supervising physician, and number of examinations repeated by a supervising physician. We considered a two-sided p value of less than .05 to be statistically significant.
Of the 402 eligible students (49% of whom were women), 193 (48%) completed the questionnaire. The respondents included 102 (53%) third-year students and 91 (47%) fourth-year students. Overall, 105 (54 %) respondents were women and 88 (46 %) were men. Of fourth-year students, 42 (46%) were women and 49 (54%) were men. Of third-year students, 57 (56%) had not yet completed the core obstetrics–gynecology rotation.
The self-reported numbers of pelvic, breast, and male genital examinations performed by fourth-year medical students are shown in Table 1. We present only the results from the fourth-year students in Table 1 because they represent the students' accumulative experiences at graduation from medical school. As expected, third-year students performed fewer examinations than did fourth-year students. We found statistically significant differences in the number of pelvic and breast examinations reported by female students as compared to those reported by male students.
The percentage of pelvic, breast, and male genital examinations reported to have been observed by a supervising physician and the percentage of exams repeated by a supervising physician for all students (third-year and fourth-year) are shown in Table 2. We combined the results for third-year and fourth-year students because supervision should be similar for all students despite level of training and combining groups strengthened the relationships found. We found significant differences between male and female students who had been observed by a supervising physician while they performed the male genital and breast examinations. The percentage of pelvic, breast, and male genital examinations that were reported to have been repeated by a supervising physician was low for all students (see Table 2).
Students were asked in the questionnaire to select if they were confident or not confident in performing pelvic, breast, and male genital examinations. The fourth-year students' responses are shown in Table 3. We present only the results from fourth-year students in Table 3 because they represent the confidence students had on graduation from medical school. As we expected, third-year students were less confident in performing the exams (data not shown).
Table 4 shows the numbers and percentages of all students (third-year and fourth-year) who were confident in performing the examinations relative to the number of examinations they had performed.
We performed a stepwise multiple regression analysis to identify factors that predict confidence in performing pelvic, breast, and male genital examinations for all students. Variables evaluated included gender, number of examinations performed, percentage of examinations observed, and percentage of examinations repeated. Only the number of examinations performed predicted confidence in performing the pelvic examination (p < .001) and breast examination (p < .001). Gender, percentage of examinations observed, and the percentage of examinations repeated by a supervising physician were not predictors of confidence for either the pelvic or breast examinations. Predictors of confidence in the male genital examination were male gender (p < .01) and number of examinations performed (p < .001).
In our study we evaluated the relationship between student gender; the number of pelvic, breast, and male genital examinations performed; the number of examinations observed; the number repeated by a supervising physician; and the students' perception of confidence in performing the examinations. Our results support our hypotheses that gender influences medical students' experiences in performing gender-specific examinations and that students with more experience are more confident. Our results did not support our hypothesis that students with more supervision are more confident. We found significant differences in the numbers of gender-specific exams performed by male and female medical students. Male medical students performed significantly fewer pelvic examinations than did female students and female students performed significantly fewer male genital examinations than did male students. This discrepancy was reflected in the student's self-described confidence in performing gender-specific exams. Nevertheless, student gender was less important in predicting confidence than was number of examinations performed. However, student gender might be considered a marker for suboptimal exposure in performing opposite-sex, gender-specific examinations.
The main predictor of confidence for both female and male students for performing the pelvic, breast, and male genital examinations was the number of examinations performed, confirming the findings of Lee et al.7 We were surprised by the absence of a supervising physician effect; student confidence did not correlate with the percentage of examinations observed or repeated by a supervising physician. Observation and repeat examination rates by supervising physicians were low, but in spite of this, a majority of students felt confident in performing breast and pelvic examinations. Although supervising physicians were more likely to observe students performing gender-based examinations on patients of the opposite sex, this did not appear to translate into a better learning experience for the student. The sheer number of examinations performed remained the most important predictor of confidence.
The percentage of pelvic, breast, and male genital examinations that were repeated by a supervising physician was low. The reasons for this are unclear. Explanations might include recall bias by the student or a sense on the part of the supervising physician that a repeated exam is unnecessary or uncomfortable for the patient. Some clinical teachers may be less comfortable supervising these aspects of the physical examination. Dixon and colleagues8 demonstrated high levels of confidence among academic general internal medicine and family medicine physicians in precepting the breast and pelvic examinations. However, supervising physicians at our institution include residents, fellows, and attending subspecialists who may feel less comfortable precepting these exams. Other reasons supervising physicians may not repeat the examination include time constraints in the ambulatory setting, being paged for more urgent matters in the inpatient setting, a report from the student that the exam is normal, and the perception that the student is not interested because the student plans to subspecialize. Whatever the reason, failure to repeat the learner's examination and confirm findings are a disservice to both the student and the patient and could pose a significant problem from a teaching and quality standpoint. It is possible that the lack of immediate feedback by an observant clinical teacher who repeats the examination contributes to a sense of overconfidence among some students. This is an area that deserves further exploration.
In our study, we evaluated only students' perceptions of confidence. Studies have shown that learners' perceived confidence does not correlate with competence.9–11 Dugoff and colleagues10 evaluated the skills of interns whom were entering obstetrics–gynecology and internal medicine and found no correlation between perceived competence and assessed performance skills. The development of competent clinical skills likely depends on the number of examinations that are performed and the verification of physical findings by the teacher. Competence may also include ease of communication with patients. It is unclear how many pelvic, breast, and male genital examinations need to be performed and appropriately supervised in order to produce true student competency in this area.
Our study had several limitations. The responses of the 48% of students who completed the questionnaire may not accurately reflect the experiences of all third-year and fourth-year students at our medical school. We did not do an analysis to assess if our respondents were representative of all students. The students who responded to the questionnaire may have had more interest in this subject or more experience performing these examinations compared to those who chose not to respond. Our study was retrospective and relied on recall that may not have been accurate. We did not specify the setting of the examinations so students may have counted examinations on plastic models as part of their clinical experience. Also, we did not provide a definition for each examination. Students may have considered Pap smears (without the bimanual examination) as a pelvic examination. Finally, we assessed only the students' confidence in performing examinations, but we did not evaluate their competence.
In summary, our study demonstrates that medical student gender influences the number of pelvic, breast, and male genital examinations performed during medical school and that student gender might be considered a marker for suboptimal exposure in performing opposite-sex, gender-specific examinations. Overall, the number of examinations observed and repeated by the supervising physician was low. Supervising physicians were more likely to observe students performing gender-based examinations on patients of the opposite sex, but this did not result in the student being more confident. The sheer number of examinations performed remained the most important predictor of confidence. Special efforts should be made to assure equal opportunities for all students to perform gender-specific examinations with direct supervision and feedback.
The authors are grateful for the assistance and statistical expertise of Doug Schaad, PhD, University of Washington. They also thank Marguerite McNeely, MD, MPH, University of Washington, for advice on questionnaire content and helpful comments.
1 Levy BT, Merchant ML. Differences in clinical experiences based on gender of third year medical students in a required family medicine preceptorship. Acad Med. 2002;77 (12 pt 1):1241–46.
2 Carney PA, Dietrich AJ, Eliassen S, Pipas C, Donahue D. Differences in ambulatory teaching and learning by gender match of preceptors and students. Fam Med. 2000;32:618–23.
3 Kann PE, Lane DS. Breast cancer screening knowledge and skills of students upon entering and exiting a medical school. Acad Med. 1998;73:904–6.
4 Wollstadt LJ, Gravdal J, Glasser M. Gender and the educational experience in a primary care training setting. Fam Med. 1990;22:210–14.
5 Blake RL Jr. Gender concordance between family practice residents and their patients in an ambulatory-care setting. 1990;65:702–3.
6 Emmons SL, Adams KE, Nichols M, Cain J. The impact of perceived gender bias on obstetrics and gynecology skills acquisition by third year medical students. Acad Med. 2004;79:326–32.
7 Lee M, Hogson CS, Wilkerson L. Predictors of self-perceived competency in cancer screening examinations. J Cancer Educ. 2002;17:180–82.
8 Dixon JG, Bogna BA, Keyserling TC, et al. Teaching women's health skills: confidence, attitudes, and practice patterns of academic generalist physicians. J Gen Intern Med. 2003;18:411–18.
9 Morgan PJ, Cleave-Hogg D. Comparison between medical students' experience, confidence and competence. Med Educ. 2002;36:534–39.
10 Dugoff l, Everett MR, Vontver L, Barley GE. Evaluation of pelvic and breast examination skills of interns in obstetrics and gynecology and internal medicine. Am J Obstet Gynecol. 2003;189:655–58.
11 Barrett SV, Zapka JG, Mazor KM, Luckmann RS. Assessing third-year medical students' breast cancer screening skills. Acad Med. 2002;77:905–10.