Dena Goffman

Simulation training improves medical students' learning experiences when performing real vaginal deliveries.

Objective: To determine the relationship between simulation training for vaginal delivery maneuvers and subsequent participation in live deliveries during the clinical rotation and to assess medical students’ performance and confidence in vaginal delivery maneuvers with and without simulation training. Methods: Medical students were randomized to receive or not to receive simulation training for vaginal delivery maneuvers on a mannequin simulator at the start of a 6-week clerkship. Both groups received traditional didactic and clinical teaching. One researcher, blinded to randomization, scored student competence of delivery maneuvers and overall delivery performance on simulator. Delivery performance was scored (1–5, with 5 being the highest) at weeks 1 and 5 of the clerkship. Students were surveyed to assess self-confidence in the ability to perform delivery maneuvers at weeks 1 and 5, and participation in live deliveries was evaluated using student obstetric patient logs. Results: Thirty-three students were randomized, 18 to simulation training [simulation group (SIM)] and 15 to no simulation training [control group (CON)]. Clerkship logs demonstrated that SIM students participated in more deliveries than CON students (9.8 ± 3.7 versus 6.2 ± 2.8, P < 0.005). SIM reported increased confidence in ability to perform a vaginal delivery, when compared with CON at the end of the clerkship (3.81 ± 0.83 versus 3.00 ± 1.0, respectively, P < 0.05). The overall delivery performance score was significantly higher in SIM, when compared with CON at week 1 (3.94 ± 0.94 versus 2.07 ± 1.22, respectively, P < 0.001) and week 5 (4.88 ± 0.33 versus 4.31 ± 0.63, P < 0.001) in the simulated environment. Conclusions: Students who receive simulation training participate more actively in the clinical environment during the course of the clerkship. Student simulation training is beneficial to learn obstetric skills in a minimal risk environment, demonstrate competency with maneuvers, and translate this competence into increased clinical participation and confidence.

Using Simulation Training to Improve Shoulder Dystocia Documentation

OBJECTIVE: To estimate whether shoulder dystocia documentation could be improved with a simulation-based educational experience. METHODS: Obstetricians at our institution (n=71) participated in an unanticipated simulated shoulder dystocia followed by an educational debriefing session. A second shoulder dystocia simulation was completed at a later date. Delivery notes were a required component of each simulation. Notes were evaluated using a standardized checklist for 16 key components. One point was awarded for each element present. Wilcoxon signed rank tests were used to compare documentation between simulations. RESULTS: Participants consisted of 43 (61%) attending and 28 (39%) resident physicians. Ages ranged from 25–63 years (mean±standard deviation 37.0±9.0), and 75% were female. Years of obstetric experience for our attendings ranged from 4 to 31 years (14.5±8.1). Documentation scores were significantly improved after training. Attendings’ baseline documentation scores were 8.5±2.2 and improved to 9.4±2.3, P=.03. Residents’ documentation scores also improved (9.0±2.1 compared with 10.6±2.2, P=.001). In particular, improvement was seen in two components of documentation: 1) providers present for shoulder dystocia (P=.007) and 2) which shoulder was anterior (P<.001). No improvement was seen in standard delivery note components (eg, date, time) or infant characteristics (eg, weight, Apgar scores). CONCLUSION: Although we showed a significant improvement in the quality of documentation through this simulation program, notes were still suboptimal. Use of standardized forms for shoulder dystocia delivery notes may provide the best solution to ensure appropriate documentation. LEVEL OF EVIDENCE: II

Improved performance of maternal-fetal medicine staff after maternal cardiac arrest simulation-based training

OBJECTIVE To determine the impact of simulation-based maternal cardiac arrest training on performance, knowledge, and confidence among Maternal-Fetal Medicine staff. STUDY DESIGN Maternal-Fetal Medicine staff (n = 19) participated in a maternal arrest simulation program. Based on evaluation of performance during initial simulations, an intervention was designed including: basic life support course, advanced cardiac life support pregnancy modification lecture, and simulation practice. Postintervention evaluative simulations were performed. All simulations included a knowledge test, confidence survey, and debriefing. A checklist with 9 pregnancy modification (maternal) and 16 critical care (25 total) tasks was used for scoring. RESULTS Postintervention scores reflected statistically significant improvement. Maternal-Fetal Medicine staff demonstrated statistically significant improvement in timely initiation of cardiopulmonary resuscitation (120 vs 32 seconds, P = .042) and cesarean delivery (240 vs 159 seconds, P = .017). CONCLUSION Prompt cardiopulmonary resuscitation initiation and pregnancy modifications application are critical in maternal and fetal survival during cardiac arrest. Simulation is a useful tool for Maternal-Fetal Medicine staff to improve skills, knowledge, and confidence in the management of this catastrophic event.

Resident training for eclampsia and magnesium toxicity management: simulation or traditional lecture?

OBJECTIVE To compare eclampsia and magnesium toxicity management among residents randomly assigned to lecture or simulation-based education. STUDY DESIGN Statified by year, residents (n = 38) were randomly assigned to 3 educational intervention groups: Simulation→Lecture, Simulation, and Lecture. Postintervention simulations were performed for all and scored using standardized lists. Maternal, fetal, eclampsia management, and magnesium toxcity scores were assigned. Mann-Whitney U, Wilcoxon rank sum and χ(2) tests were used for analysis. RESULTS Postintervention maternal (16 and 15 vs 12; P < .05) and eclampsia (19 vs 16; P < .05) scores were significantly better in simulation based compared with lecture groups. Postintervention magnesium toxcitiy and fetal scores were not different among groups. Lecture added to simulation did not lead to incremental benefit when eclampsia scores were compared between Simulation→Lecture and Simulation (19 vs 19; P = nonsignificant). CONCLUSION Simulation training is superior to traditional lecture alone for teaching crucial skills for the optimal management of both eclampsia and magnesium toxicity, 2 life-threatening obstetric emergencies.

Maternal morbidity and risk of death at delivery hospitalization.

OBJECTIVE: To examine the effect of underlying maternal morbidities on the odds of maternal death during delivery hospitalization. METHODS: We used data that linked birth certificates to hospital discharge diagnoses from singleton live births at 22 weeks of gestation or later during 1995–2003 in New York City. Maternal morbidities examined included prepregnancy weight more than 114 kilograms (250 pounds), chronic hypertension, pregestational or gestational diabetes mellitus, chronic cardiovascular disease, pulmonary hypertension, chronic lung disease, human immunodeficiency virus (HIV), and preeclampsia or eclampsia. Associations with maternal mortality were estimated using multivariate logistic regression. RESULTS: During the specified time period, 1,084,862 live singleton births and 132 maternal deaths occurred. Patients with increasing maternal age, non-Hispanic black ethnicity, self-pay or Medicaid, primary cesarean delivery, and premature delivery had higher rates of maternal mortality during delivery hospitalization. From the entire study population, 4.1% had preeclampsia or eclampsia (n=44,004), 1.8% had chronic hypertension (n=19,647), 1.1% of patients were classified as obese (n=11,936), 0.7% had pregestational diabetes (n=7,474), 0.4% had HIV (n=4,665), and 0.01% had pulmonary hypertension (n=166). Preeclampsia or eclampsia (adjusted odds ratio [OR], 8.1; 95% confidence interval [CI], 5.5–12.1), chronic hypertension (adjusted OR, 7.7; 95% CI 4.7–12.5), underlying maternal obesity (adjusted OR, 2.9; 95% CI 1.1–8.1), pregestational diabetes (adjusted OR, 3.3; 95% CI 1.3–8.1), HIV (adjusted OR, 7.7; 95% CI 3.4–17.8), and pulmonary hypertension (adjusted OR, 65.1; 95% CI 15.8–269.3) were associated with an increased risk of death during the delivery hospitalization. CONCLUSION: The presence of maternal disease significantly increases the odds of maternal mortality at the time of delivery hospitalization. LEVEL OF EVIDENCE: II

Improved obstetric safety through programmatic collaboration.

Healthcare safety and quality are critically important issues in obstetrics, and society, healthcare providers, patients and insurers share a common goal of working toward safer practice, and are continuously seeking strategies to facilitate improvements. To this end, 4 New York City voluntary hospitals with large maternity services initiated a unique collaborative quality improvement program. It was facilitated by their common risk management advisors, FOJP Service Corporation, and their professional liability insurer, Hospitals Insurance Company. Under the guidance of 4 obstetrics and gynecology departmental chairmen, consensus best practices for obstetrics were developed which included: implementation of evidence based protocols with audit and feedback; standardized educational interventions; mandatory electronic fetal monitoring training; and enhanced in-house physician coverage. Each institution developed unique safety related expertise (development of electronic documentation, team training, and simulation education), and experiences were shared across the collaborative. The collaborative group developed robust systems for audit of outcomes and documentation quality, as well as enforcement mechanisms. Ongoing feedback to providers served as a key component of the intervention. The liability carrier provided financial support for these patient safety innovations. As a result of the interventions, the overall AOI for our institutions decreased 42% from baseline (January-June 2008) to the most recently reviewed time period (July-December 2011) (10.7% vs 6.2%, p < 0.001). The Weighted Adverse Outcome Score (WAOS) also decreased during the same time period (3.9 vs 2.3, p = 0.001.) Given the improved outcomes noted, our unique program and the process by which it was developed are described in the hopes that others will recognize collaborative partnering with or without insurers as an opportunity to improve obstetric patient safety.

Simulation in maternal-fetal medicine: making a case for the need.

Medical error remains a cause for concern in obstetrics. Studies have shown that the most important contributing factors to adverse events in obstetrics often relate to poor teamwork and ineffective communication. A potential solution to these problems includes transforming institutions, obstetric departments, and maternal-fetal medicine divisions into high-reliability organizations. Simulation is a valuable tool which can be utilized, in obstetrics and maternal-fetal medicine, as an integral part of programs designed to advance knowledge and technical skills; improve communication and team function; and identify and correct systems issues. Simulation should be an integral part of our journey towards high reliability with the ultimate goal of improving patient safety and quality of care in obstetrics.

Simulation as a tool for improving acquisition of neonatal resuscitation skills for obstetric residents

Abstract Objective: Our goal was to compare the confidence, knowledge, and performance of obstetric residents taught initial neonatal resuscitation steps in a simulation-based versus lecture-based format. Methods: Our study was a prospective randomized controlled trial of 33 obstetric residents. Baseline confidence, knowledge, and clinical skills assessments were performed. Subjects were randomized to traditional lecture (n = 14) or simulation-based (n = 19) neonatal resuscitation curriculum with a focus on initial steps. Follow-up assessments were performed at 3 and 6 months. Total confidence, knowledge, and clinical performance scores and change from baseline in these scores were calculated and compared between groups. Results: Both the lecture-based and simulated-based groups demonstrated significant improvement in confidence, knowledge, and performance over time. However, compared with the lecture group, the magnitude of the mean change from baseline in performance scores was significantly greater in the simulation group at 3 months (2.9 versus 10.1; p < 0.001), but not at 6 months (7.0 versus 9.3; p = 0.11). Conclusions: Our study demonstrates the superiority of simulation in teaching obstetric residents initial neonatal resuscitation steps compared with a traditional lecture format. Skills are retained for upwards of 3–6 months. Refresher instruction by 6 months post-instruction may be beneficial.

Retention of skills 2 years after completion of a postpartum hemorrhage simulation training program in rural Rwanda

To evaluate the long‐term retention of skills gained by rural physicians who completed a postpartum hemorrhage simulation‐training program.

Clinical Outcomes of Anterior Compared With Posterior Placenta Accreta

INTRODUCTION: The objective of this study was to evaluate the effect of placental location on maternal and gestational outcomes associated with placenta accreta. METHODS: We conducted a retrospective analysis of all cases of pathologically proven placenta accreta with delivery after 20 weeks of gestation at our institution over the past 10 years. Placental location was characterized as anterior or nonanterior. Central previa location was considered in the anterior group. Gestational age at delivery and maternal data were collected. Outcomes of patients diagnosed with anterior accretes were compared with those of patients diagnosed with nonanterior accretes. Composite maternal morbidity was defined as nonred blood cell product transfusion, disseminated intravascular coagulation, or intensive care unit admission or postpartum stay longer than 7 days. RESULTS: The overall incidence of placenta accreta was 1.2 per 1,000 deliveries with 51 confirmed cases. Thirty-one (60.8%) had anterior and 20 (39.2%) had nonanterior placentation. Patients with anterior accretes were more likely to be antenatally diagnosed (58% compared with 17%, P=.003), had a history of more cesarean deliveries (P=.04), and deeper invasion of chorionic villi (P=.004). Composite maternal morbidity was significantly higher (71% compared with 30%, P=.007) and mean gestational age at delivery was lower (33.3 compared with 36.6 weeks, P=.02) for patients with anterior placenta accreta. CONCLUSION: Nearly 40% of accreta cases at our academic institution had nonanterior location. These cases are more difficult to diagnose antenatally but also seem to have less associated morbidity. Anterior placenta accreta appears to be a more histologically and clinically severe disease than posterior placenta accreta with significantly higher maternal morbidity and preterm delivery despite improved antenatal detection.