Materials and Methods

We conducted a prospective, randomized, controlled trial of medical students at Alpert Medical School of Brown University (Providence, RI) who were rotating through their obstetrics and gynecology clerkship from November 2007 to May 2008. Our student population rotating through obstetrics and gynecology includes both third- and fourth-year students.

Students at our institution may delay completion of core clerkships until their fourth year to complete elective rotations based on their interests. Preclinical exposure to pelvic anatomy and pelvic floor dysfunction at our institution includes an introductory lecture on the pelvic floor muscles prior to pelvic anatomy dissection during gross anatomy in the fall of their first year of medical school.

All students completing their obstetrics and gynecology clerkship during the study period were eligible. This study was approved by the Institutional Review Board at Women and Infants’ Hospital (Providence, RI).

A detailed description of the study was provided to the students on the first day of each clerkship, and consent was obtained from interested students. Students completed baseline knowledge and attitude questionnaires at the time of consent and at the end of their clerkship. Students were randomized by clerkship using the Moses-Oakford algorithm and a random numbers table to receive either usual teaching and a 1 hour session with the trainer (described in the following text) vs usual teaching alone.

Usual teaching for female PA and PFD during the 6 week obstetrics and gynecology clerkship consists of 2 weeks of gynecologic surgery, 1 lecture on urinary incontinence, and a half-day of outpatient urogynecology. We chose to randomize the students by clerkship to minimize the potential for many social interaction threats within a clerkship, such as diffusion, which may occur if students in the trainer group discuss the intervention with students in the usual teaching group on the same clerkship; compensatory rivalry and reactivity among students in the same clerkship; and resentful demoralization, which would be detrimental to students in the usual teaching group. We believe that these interactions would have threatened the internal validity of our study and could have created significant anxiety among students that could adversely affect the learning environment

Computer pelvic trainer intervention

We developed an interactive computer pelvic trainer (trainer) for individual-based learning of female PA and PFD, consisting of a 3-dimensional, interactive female pelvis and 4 teaching modules using dynamic animation including: (1) anatomy, (2) PFD fundamentals, (3) history, and (4) physical examination.

The 3-dimensional interactive pelvis uses embedded animation and can be navigated by the user in virtual space. It includes aspects such as user-controlled points of view, transparency texturing, roll-over naming, and virtual dissection ( Figure 1 ). The pelvic anatomy scaffold for the pelvis was developed using custom 3-dimensional models of the female pelvis with digitized images of the Visible Human Project acquired from the National Library of Medicine, computed tomography, and 3-dimensional ultrasound.

Interactive pelvis with uterus highlighted and named with rollover function

Hampton. Improving medical student knowledge of female PFD and anatomy. Am J Obstet Gynecol 2010.

Three-dimensional modeling and Maya animation software (Autodesk, Inc., San Rafael, CA) were used to resurface this scaffold, allowing for enhanced texture mapping and hyperrealistic display. The learner can navigate and manipulate the interactive pelvis at their own pace and refer back to it at any time during the trainer session. Such navigation and manipulation in real time allows for true 3-dimensional understanding vs observing video or still pictures.

The 4 teaching modules utilize voice-over, embedded animation and video clips to emphasize anatomy, the fundamentals of pelvic floor dysfunction, and how to perform a targeted history and physical examination. On average, the 4 modules can be completed in 30 minutes. They can be self-navigated with on-screen controls, and the learner can control their learning experience by focusing on certain modules, repeating modules or concepts, and referring back to learning points.

The anatomy module is a guided tour of a dynamic, 3-dimensional pelvis emphasizing the structures important to pelvic organ support. The fundamentals module teaches basic principles of urinary incontinence, including stress and urge incontinence, and their potential etiologies. Principles of pelvic organ prolapse are discussed and different degrees and types of pelvic organ prolapse reviewed, with teaching enhanced by integrating computer animations of pelvic anatomy.

The history taking module includes a video presentation of a thorough urogynecologic history-taking session with physician and patient and is edited so that important history-taking concepts are highlighted and emphasized, such as questions targeting incontinence and prolapse symptoms, and the impact these disorders may have on a woman’s quality of life. The physical examination module includes a video presentation of a thorough urogynecologic pelvic examination. Examples of pelvic organ prolapse are presented in split-screen format of real video footage and dynamic computer animation, with the anatomy changing as pressure is placed on the pelvic floor ( Figure 2 ). The viewer sees the anatomic finding from different viewpoints, with organs transparent, and with specific emphasis on anatomic mechanisms contributing to the defect.

Physical exam module with split screen

Hampton. Improving medical student knowledge of female PFD and anatomy. Am J Obstet Gynecol 2010.

The interactive pelvis and the 4 modules were designed to improve medical student knowledge in the following domains as dictated by APGO medical student educational objectives: (1) female pelvic anatomy and (2) diagnosis of female pelvic floor dysfunction through history and physical examination. Specifically, incorporating screening questions for incontinence into a history, differentiating between types of incontinence, obtaining a pertinent history for incontinence, and identifying pelvic floor defects on physical exam are priority 1 and 2 objectives.

This trainer was developed with the support from an educational grant from the Association of Professors of Gynecology and Obstetrics.

Students randomized to the trainer group were assigned to a 1 hour monitored session with the trainer during clerkship self-study time. Students were compensated at the completion of the study. At the conclusion of the study, all students had the opportunity to access the trainer.

Results

After randomization, the first 2 clerkships of the study period were randomized to usual teaching and the second 2 clerkships were randomized to the intervention. Forty-three students were enrolled in the study (21 randomized to intervention; 22 randomized to usual teaching). All students completed participation and pre- and postintervention questionnaires. All students randomized to the trainer were able to complete all modules during their trainer session, and the time needed to complete the trainer ranged from 45 to 60 minutes.

At baseline, the 2 student groups were similar in the number of third- vs fourth-year students (100% vs 96%, respectively; P > .05). The trainer group had a higher proportion of students exposed to surgery (67% vs 32%), urology (5% vs 0%), and women’s health (20% vs 0%) compared with the usual teaching group. At baseline, there was no difference in preintervention scores between groups for either knowledge scores (mean score, 10.3 ± 2.3 vs 10.2 ± 2.7 for trainer vs usual teaching, respectively; P = .9) or attitude scores (mean score, 9.1 ± 2.1 vs 8.1 ± 2.4 for trainer vs usual teaching, respectively; P = .15).

Postintervention knowledge scores were significantly higher in students randomized to the trainer compared with usual teaching (mean score, 15.6 ± 1.9 vs 12.6 ± 2.5 for pelvic trainer vs usual teaching, respectively; P = .0007). When we evaluated specific APGO knowledge domains, we found that students randomized to the trainer had significantly higher postintervention knowledge of anatomy (mean score, 2.9 ± 1.1 vs 1.7 ± 0.8; P = .0005) and types of urinary incontinence (mean score, 4.0 ± 1.0 vs 3.2 ± 1.0; P = .009). Both groups improved in the remaining 3 knowledge domains (screening, history taking, and physical examination), with no differences between groups.

Postintervention attitude scores were significantly higher in students randomized to the trainer compared with usual teaching (mean score, 19.2 ± 2.8 vs 15.8 ± 3.2 for trainer vs usual teaching, respectively; P = .001), indicating a higher level of comfort with their knowledge regarding female PA and the diagnosis of PFD. On multiple linear regression, students randomized to the trainer had on average a 2 point (± .81) higher mean knowledge score and a 2.7 point (± 1.1) higher mean attitude score compared with students randomized to usual teaching, after adjusting for year of medical education and completion of prior relevant clerkship ( P = .02 for both).

Finally, we evaluated trainer-specific satisfaction and usability questions for students randomized to the trainer. Ninety-five percent of students felt that the trainer was easy to navigate. All students (100%) reported they felt it improved their knowledge of female PFD and 95% felt it improved their knowledge of female PA. Ninety-five percent of students who used the trainer felt that its use would lead to improved care of their patients in the future, and 81% thought that using the trainer would help them on their obstetrics and gynecology shelf examination. Ninety percent would use it on their own if they were given access to it, and 75% would have liked to have used the trainer for longer than the assigned 1 hour session.

Comment

The field of female pelvic floor disorders, whose foundation rests on the basic framework of pelvic anatomy, is becoming increasingly important as our population ages. We developed a novel, computer-based learning trainer that improves both knowledge and attitudes of medical students regarding female PFD and PA and can be easily integrated into an obstetrics and gynecology clerkship curriculum.

As learners begin to fall into the net generation group, teaching formats need to change to meet evolving methods of learning and as such should embrace nonlinear, personalized approaches to learning. In addition, successful adult education should be learner centered and active rather than passive. The trainer used in this study enables active, learner-centered, nonlinear learning by allowing students to navigate through the modules as desired, individualizing their learning experience by spending more or less time on modules and interacting with the 3-dimensional pelvis as needed. With the increased demands on both learner and teacher time, the ability to use the trainer as an individual learning tool is increasingly important.

Educators also need to make decisions regarding the compatibility of the topic they are teaching with the media being used to teach it because this may influence its effectiveness. Student learning and recall are positively affected by visual material, and the trainer in this study emphasizes the visual aspects of learning. In contrast to previous educational tools, the trainer is updated, with hyperrealistic novel animations and 3-dimensional visualization.

Our study findings are consistent with previous studies showing that pelvic anatomy knowledge can be improved using computer-based learning tools; however, our trainer is unique because it is on an interactive platform and includes pelvic floor dysfunction teaching. Corton et al compared computer-based learning, using structured text and illustrations with color overlays, links to still patient photos, cadaver dissection videos, and animations, to conventional format text-based instruction for student, resident, and fellow learning of pelvic anatomy. Knowledge scores improved with both formats, but satisfaction was highest with computer-based learning.

The study by Corton et al included 12 students and did not include pelvic floor dysfunction. Sultana et al compared CD-ROM, using still photographs and video with voice-over, and video, using still drawings and photographs with voice-over, for third-year medical student review of pelvic anatomy. The authors found that the CD-ROM was associated with improved knowledge and acceptability in students for pelvic anatomy but did not include pelvic floor dysfunction.

There are limitations to our study. First, we were able to compare only composite scores between groups to maintain student anonymity and a healthy learning environment. This was done to ensure student participation in the study while minimizing any anxiety students may have that performance on study questionnaires would influence grading in the clerkship. Although we likely avoided social interaction threats, a Hawthorne effect may still have occurred for the group that was assigned to the trainer. Also, we collected information on previous general surgery, urology, and women’s health clerkship completion; however, we did not collect information on the completion of family medicine or internal medicine clerkships at the time of enrollment.

Another potential limitation is that although we developed our knowledge and attitude questionnaires based on APGO learning objectives for medical students and pilot-tested them prior to use, they have not been previously used in other studies. However, at the start of our study, we were unable to identify any previously validated and published questionnaires testing the domains in both female PA and PFD targeted by our trainer. In addition, we were unable to evaluate whether the trainer improved retention of knowledge or changed ultimate practice patterns compared with usual teaching.

Finally, it is unclear whether improved knowledge and attitudes regarding female PA and PFD may lead to improved care for women in the future. However, it is our belief that improved medical student attitudes and knowledge at the end of their obstetrics and gynecology clerkship is important and beneficial for women’s health and for physicians caring for women in any field in the future.

The trainer is a sophisticated and feasible tool utilizing appropriate technology to teach a complex, 3-dimensional topic and can be an important adjunct to learning. Supported by an educational grant from the Association of Professors of Gynecology and Obstetrics, this trainer can be incorporated into an obstetrics and gynecology medical student curriculum. Such implementation, in a background of limited teaching time, can improve knowledge and attitudes about female PA and the diagnosis, management, and counseling of female PFD.