Holmström, Shelly W. MD; Downes, Katheryne MPH; Mayer, James C. MD; Learman, Lee A. MD, PhD
From the Division of General Obstetrics and Gynecology, the Office of Clinical Research, and the Division of Reproductive Endocrinology, University of South Florida, Tampa, Florida; and the Division of General Obstetrics and Gynecology, Indiana University, Bloomington, Indiana.
Presented at the CREOG and APGO annual meeting, March 9–12, 2011, San Antonio, Texas.
Corresponding author: Shelly W. Holmström, MD, Department of Obstetrics and Gynecology, 2 Tampa General Circle, 6th Floor, Tampa, FL 33606; e-mail: email@example.com.
Financial Disclosure The authors did not report any potential conflicts of interest.
Simulation-based training is being used more frequently in many medical specialties and has been used for many years in the military and airline industries. Medical simulations attempt to recreate clinical scenarios that are often complex and important to understand. Many scenarios involve events that are rare but potentially life-threatening.1 Simulation training allows learners to make mistakes in a controlled setting and to repeat the simulation or skills multiple times.1,2
Simulation training can be used to teach medical students and physicians in a safe environment without risk to a patient. Obstetrics simulators have been used to teach rare and catastrophic events to improve patient safety and improve the competency of the learners. The National Capital Consortium Residency program has developed a simulation-based curriculum to train their resident physicians. This curriculum includes simulation in spontaneous vaginal delivery, perineal laceration repair, McDonald cerclage, assisted vaginal delivery, low forceps delivery, cesarean delivery, hysterectomy, and laparoscopy.1 The faculty of this residency program has demonstrated simulation training improves resident physician performance of term vaginal breech delivery,3 shoulder dystocia,4 and documentation of shoulder dystocia events.5
A literature search on PubMed demonstrated four previous studies describing the use of an obstetrics simulator in undergraduate medical education. Jude et al6 reported an increase in medical student confidence with vaginal deliveries after simulation training. Dayal et al7 showed that students who received simulation training had more confidence to perform vaginal delivery maneuvers and participated in more live vaginal deliveries. Deering et al8 demonstrated a significant improvement in medical students' perceived comprehension and comfort with the performance of several common obstetric procedures. Reynolds et al9 showed that a simulated vaginal delivery improved short-term knowledge and satisfaction. These previous studies examined a smaller number of students and did not evaluate the effect of simulation on the students' final examination and evaluation scores.
The purpose of this study was to assess confidence in and active participation in vaginal deliveries and to compare the end of clerkship examination scores during the University of South Florida's third-year medical school maternal newborn clerkship.
MATERIALS AND METHODS
Medical students at the University of South Florida have a horizontally integrated curriculum during their third year of education. Students learn obstetrics and gynecology during three separate clerkships: surgical care, primary care and special populations, and maternal newborn. Two of the 8 weeks of the surgical care clerkship are devoted to gynecologic surgery. One month of the 3 months of the primary care clerkship is devoted to women's outpatient health care. The maternal newborn clerkship is 4 weeks in length and consists of labor and delivery, outpatient obstetrics, and newborn nursery experiences. Eight to 10 students rotate through the maternal newborn clerkship each month. All students participate in vaginal and cesarean deliveries at Tampa General Hospital.
Maternal newborn medical students rotating from February 1, 2010 to January 31, 2011 were invited to participate in this research study. This timeframe included 12 different clerkship groups. An average of 120 medical students is enrolled in the maternal newborn clerkship during a 1-year time period. As a convenience sample, this study enrolled every medical student who consented to this research during the 1 year of study. The clerkship groups were randomized as a group to participate in a lecture on vaginal deliveries plus a simulated vaginal delivery or lecture only (Fig. 1). The clerkship group's assignment to “simulation” or “no simulation” was sealed in an opaque envelope and was randomly selected by the primary author (S.W.H.) at the start of each new clerkship. Students were invited to participate during their orientation to this clerkship and to sign an informed consent form. The University of South Florida Institutional Review Board approved this study before enrolling any students.
All students completed a baseline questionnaire assessing their baseline knowledge of vaginal deliveries, confidence to perform various portions of a delivery, and previous experience with vaginal deliveries before their group assignment was revealed. Next, all students attended a formal lecture on a vaginal delivery, which included viewing a 5-minute vaginal delivery video (Brookside Associates). The control group of students then repeated the self-assessment questionnaire immediately after the lecture to assess their confidence to perform maneuvers involved in a vaginal delivery. This questionnaire was based on a previously validated evaluation tool created by Jude et al6 and included an assessment of the student's ability to define the stages of labor, to deliver the fetal head, shoulders, and body, to deliver and examine the placenta, and to determine the extent of lacerations. Levels of graduated responsibility were used to assess the students' confidence. Students rated their ability to perform various portions of the vaginal delivery as “not at all,” “with close hand-on-hand supervision,” “with minimal supervision,” or “independent, with back-up for problems.”
The study group of students participated in simulated vaginal deliveries after the formal lecture. These students were divided into two groups of four to six students and had a preceptor-led teaching and practice session on a Noelle obstetrics simulator (Gaumard Scientific) at the University of South Florida simulator center. The instructor demonstrated a simulated vaginal delivery and allowed each student to perform at least one simulated vaginal delivery. The students were supervised by one of authors (S.W.H.) during all of the simulated vaginal deliveries and were given corrective feedback on delivery maneuvers. The total time to train each small group of students was approximately 60 minutes. Study group students then completed the questionnaire to assess their confidence to perform vaginal delivery maneuvers.
After the questionnaire was completed a second time by both the control and study groups, all students received a packet of three questionnaires to complete after each of their first three vaginal deliveries. All students then completed the questionnaire for the fourth and final time on the last day of the month-long clerkship.
The final written examination, oral examination, evaluation scores, and number of vaginal deliveries in which the students participated were collected at the conclusion of each clerkship. The oral examination was administered during the last week of the clerkship by a panel of three physicians, one of whom was the simulation instructor (S.W.H.). Faculty and resident physicians evaluating the students during the maternal newborn clerkship were blinded to the randomization groups. The responses to the questionnaire as well as these other measures of assessment were compared between the control and study groups.
The students' demographics, baseline characteristics, and confidence to perform vaginal delivery maneuvers with or without the simulation training were analyzed. Student responses of “not at all” or “with close hand-on-hand supervision” represented lack of confidence and were compared with “with minimal supervision” or “independently,” which represented confidence. Group comparisons were performed using the χ2, Fisher exact, and Mann-Whitney U test; P<.05 was considered statistically significant.
One hundred thirteen medical students (94.9% of those eligible) participated in this research study. The two groups of medical students participating in this study did not differ significantly by gender or age. The study group of medical students who underwent simulation training consisted of 57 students (57.9% women) and the control group who did not undergo simulation training consisted of 56 students (48.2% women) (χ2, P=.303). The median and range of age was 25 years (23–45) for the simulation group and 25 years (22–37) for the lecture-only group (Mann-Whitney U, P=.752). The two groups of students were equally confident on vaginal delivery maneuvers at baseline, with the exception that the simulator trained students were somewhat more confident that they were ready to attempt a delivery with attending assistance (P=.042) as compared with the nonsimulator-trained students (Table 1).
On the immediate postintervention self-assessment survey, students receiving simulation training were significantly more confident in their ability to perform a vaginal delivery immediately after assessment than students who did not receive the simulation training (P<.01). The simulator-trained students reported a higher level of confidence that they could control the fetal head (P<.001), deliver the shoulders (P<.001), deliver the fetal body after the shoulders (P<.001), deliver the placenta (P<.001), examine the placenta (P<.001), define the extent of the lacerations (P=.032), and participate in deliveries (P<.001) compared with nonsimulator-trained students (Table 2). There was no statistically significant difference in the confidence to attempt a full vaginal delivery alone (P=.113) between the two groups. Interestingly, the nonsimulator-trained students were more confident that they could define the stages of labor after the lecture only (P=.047) as compared with the simulator-trained students.
When asked if they felt confident that they could perform a delivery with minimal supervision or independently with attending assistance, 30 students (52.6%) in the simulator-trained group reported this level of confidence as compared with nine students (16.1%) in the nonsimulator-trained group (P<.001). Similarly, 29 students (50.9%) in the simulator-trained group were confident that they could perform a delivery with minimal supervision or independently with resident assistance as compared with eight students (14.3%) in the nonsimulator-trained group (P<.001). The two groups of students did not differ significantly in the number of students reporting a sufficient level of confidence to attempt a delivery alone. Six of the 57 students (10.5%) in the simulator-trained group and one of the 56 students (1.8%) reported confidence to perform a delivery independently (P=.113).
The increased level of confidence was not apparent on subsequent survey responses from the two groups of students. There was no statistically significant difference in the number of students reporting confidence to perform the vaginal delivery maneuvers between the two groups on nearly every question completed after each of their first three deliveries. The simulation group demonstrated a higher level of confidence on almost all areas on the final survey completed at the conclusion of the clerkship, but this difference failed to reach statistical significance (Table 3).
Simulator-trained students scored significantly higher on their oral (P=.004) and written examinations (P=.009) 4 weeks after the intervention (Table 4). The two groups of students did not report a statistically significant number of vaginal deliveries in which they participated (P=.203) and did not differ in their final evaluation scores from the resident and attending physicians (P=.966).
The results of this study show that medical students who participated in a simulated vaginal delivery early in the maternal newborn clerkship had higher levels of self-assessed confidence to perform obstetric maneuvers immediately after the intervention. The two groups of students did not display significantly different levels of confidence with vaginal delivery maneuvers at the conclusion of the clerkship. However, post hoc power analysis indicated that between 165 and 10,000 participants per group would have been necessary to find statistically significant differences between the two groups on the final survey. This finding is most likely the result of a “ceiling effect” in that, by the final survey, all of the students were reporting a high level of confidence (90% or more on most questions).
The simulator-trained students may been more invested in the maternal newborn clerkship and gained confidence to participate in a delivery more quickly as a result of the simulation activity, which may explain why these students scored higher on their oral and written examinations. These conclusions are supported by a study conducted by Nackman et al10 during a required fourth-year surgical clerkship. These authors observed that a simulator session with a computerized life-size mannequin improved medical students' scores on an objective structured clinical examination conducted at the end of the clerkship. This relationship between simulation training and examination scores would make an intriguing topic for future research.
The primary strengths of this study are the number of participating students and the randomized controlled design. The authors were able to enroll nearly every eligible medical student during the 12-month period of this project. Only one student declined participation at the time of orientation and the other five were excluded because they were not present at orientation. The students were randomly assigned to participate in the simulated vaginal delivery or not based on which clerkship group they were assigned. Our medical students do not rotate through all their third-year clerkships with the same group of students throughout the year. The students are able to independently select the order in which they rotate before the beginning of the third year of medical school. This self-selected schedule further contributes to the random nature of this study because some students had already gynecologic surgical or outpatient experiences in the surgical care clerkship or the primary care and special population clerkship, or both, whereas others had not.
This study does have several limitations. The 1 hour of instruction with an attending physician during simulation training and the feedback that the students received at the time of simulation may have improved their level of confidence independent of the obstetrics simulation session. Additionally, a measurement of earlier participation in a live vaginal or cesarean birth could not be calculated between the two groups because some of the students are not assigned to labor and delivery until the third week of the clerkship because of the inherit structure of the clerkship. Moreover, only one attending physician (S.W.H.) gave the lecture and oversaw the simulation session. This same attending was also one of the oral examiners, which could have potentially biased the study. This project was limited to a single institution, which has a unique third-year curriculum design. Our students experience obstetrics and gynecology during three clerkships during their third year of study: maternal newborn, surgical care and primary care, and special populations. The authors are not aware of any other medical schools in the United States that have the same curricular design and, thus, the results may not be generalized to all other medical schools.
Simulation training is being used to a larger extent in undergraduate and graduate medical education. Providing adequate time and resources for this type of training is critical. Simulation is ideally suited for the adult learner and has great potential for the acquisition of obstetric procedures. Medical students and resident physicians can perform interventions and procedures without fear of harming an actual patient and can receive constructive feedback in a safe environment. Hopefully, an increasing amount of simulation-based curriculum will be incorporated into the education of medical students.
1. Macedonia CR, Gherman RB, Satin AJ. Simulation laboratories for training in obstetrics and gynecology. Obstet Gynecol 2003;102:388–92.
2. Johannsson H, Ayida G, Sadler C. Faking it? Simulation in the training of obstetricians and gynaecologists. Curr Opin Obstet Gynecol 2005;17:557–61.
3. Deering S, Brown J, Hodor J, Satin AJ. Simulation training and resident performance of singleton vaginal breech delivery. Obstet Gynecol 2006;107:86–9.
4. Deering S, Poggi S, Macedonia C, Gherman R, Satin AJ. Improving resident competency in the management of shoulder dystocia with simulation training. Obstet Gynecol 2004;103:1224–8.
5. Deering S, Poggi S, Hodor J, Macedonia C, Satin AJ. Evaluation of residents' delivery notes after a simulated shoulder dystocia. Obstet Gynecol 2004;104:667–70.
6. Jude DC, Gilbert GG, Magrane D. Simulation training in the obstetrics and gynecology clerkship. Am J Obstet Gynecol 2006;195:1489–92.
7. Dayal AK, Fisher N, Magrane D, Goffman D, Bernstein PS, Katz NT. Simulation training improves medical students' learning experiences when performing real vaginal deliveries. Simul Healthc 2009;3:155–9.
8. Deering SH, Hodor JG, Wylen M, Poggi S, Nielsen PE, Satin AJ. Additional training with an obstetric simulator improves medical student comfort with basic procedures. Simul Healthcare 2006;1:32–4.
9. Reynolds A, Ayres-de-Campos D, Bastos LF, van Meurs WL, Bernardes J Effect of labor and delivery simulation classes in undergraduate medical learning. Med Educ Online 2008;13:14. Available at: http://www.med-ed-online.org
. Retrieved November 9, 2010.
10. Nackman GB, Bermann M, Hammond J. Effective use of human simulators in surgical education. J Surg Res 2003; 115:214–8.
© 2011 by The American College of Obstetricians and Gynecologists.