Simulation is becoming an integral part of the training and assessment of obstetricians. Given the variety of manual skills that must be learned, awake patients and high-risk environment, obstetrics is uniquely suited for simulation. Simulation provides opportunities to rehearse and learn from mistakes without risks to patients. The use of simulation can help overcome some limitations of the current medical education and practice environment, including work-hour limitations and concerns for patient safety. Both low- and high-fidelity simulation models can be used to accomplish educational goals. Basic and advanced skills as well as the management of obstetric emergencies are amenable to simulation. For a simulation programme to be successful, one must identify the learner and the skills that are to be learned. In the future, simulation will be more available and realistic and will be used not only for education, but also for ongoing assessment of providers.
The evolving interest in simulations for obstetrics stems from contemporary changes in medical education and concerns for patient safety. Core reasons for using simulation in a medical setting include initial learning of new skills, maintenance of already learned skills and demonstration of proficiency in required skills.
Initial learning of skills
Medical training has long relied on the apprenticeship model. This is a logical and effective way to learn; however, in the context of learning medical care, it does not completely obviate the risk of harm to patients. Even though learners are initially supervised during their first patient encounters, their hands are on the patient and their actions may cause injury before the teacher could intervene.
Simulation allows learners to practise new hands-on skills in an environment that puts no patient at risk. When the learner knows that a mistake will hurt no one, they may be more likely to act quickly and learn from mistakes. Unlike clinical situations, a simulation can be paused, stopped or restarted to correct mistakes and make teaching points. In addition, in obstetrics, our patients are usually awake and it is obvious to them when a learner is being taught how to perform a procedure during their delivery, surgery or ultrasound. Moving the initial teaching of new procedures from the bedside to the simulator can provide comfort to all involved.
Shorter hospital stays, more outpatient care and mandated reduction in work hours have resulted in less availability of patients for teaching. In the United States, professional liability concerns may prompt attending physicians to quickly take over care from a resident in an emergency, further decreasing trainees’ opportunity to learn correct management. Some of this reduction in patient exposure can be supplemented by simulation.
Maintenance of skills
Medical professionals learn innumerable skills during their training, and as a specific vocation is chosen, the breadth of skills is narrowed, but the required depth of knowledge and proficiency increases. Certain skills are performed on such a regular basis, such as a routine spontaneous vaginal delivery, that habit and muscle memory take over and little thought is needed to perform the task correctly and safely. Other skills, such as managing a shoulder dystocia, occur infrequently enough that providers will likely never acquire, much less maintain, a high level of performance. In these cases, simulation can provide a safe setting to practise and develop correct habits.
Maintenance of skills
Medical professionals learn innumerable skills during their training, and as a specific vocation is chosen, the breadth of skills is narrowed, but the required depth of knowledge and proficiency increases. Certain skills are performed on such a regular basis, such as a routine spontaneous vaginal delivery, that habit and muscle memory take over and little thought is needed to perform the task correctly and safely. Other skills, such as managing a shoulder dystocia, occur infrequently enough that providers will likely never acquire, much less maintain, a high level of performance. In these cases, simulation can provide a safe setting to practise and develop correct habits.
Testing and demonstration of proficiency
As discussed above, certain skills are performed so often that a new learner can be observed a sufficient number of times to ensure their proficiency. For rare events, a trainee might be exposed to such a small number of events that is not possible to certify them as competent. Simulation can provide a platform for teaching and testing in these settings.
For early learners, such as medical students, simulation can play a role in testing performance of basic skills. American medical schools have incorporated an objective structured clinical exam (OSCE) as part of the medical student clerkship, testing students on basic skills using simulation or patient-actors.
There are a number of skills that any practising obstetrician should be able to manage (e.g., postpartum haemorrhage, vaginal breech delivery and eclampsia); simulation can be used in these cases to test a provider and ensure that they have a basic understanding of the required skills. Hospital risk managers, malpractice insurance providers and national certification bodies are beginning to use demonstration of skills by simulation for credentialling, insurance coverage and certification. Physicians who have taken time away from clinical obstetrics, for example, to raise a child or pursue an advanced degree, may want to re-enter the field. Simulation can be used as part of the re-entry process, allowing providers to refresh their manual skills and decision-making skills under stress.
What types of simulators are available for obstetrics?
For many years, a doll and bony pelvis have been used to teach the cardinal movements of labour and delivery. From this simple concept, groups have developed more complex models and commercial products.
Simulation models are often classified by their resemblance, or fidelity, to the actual patient and event that they are trying to reproduce. Models are typically classified as either low or high fidelity. Low-fidelity models, such as the pelvis and doll described above, are typically simple, inexpensive to make or purchase and attempt to reproduce a narrow portion of reality. Low-fidelity simulators typically require an instructor to directly interact with the learner to guide and narrate the scenario. High-fidelity simulators, on the other hand, attempt to reproduce reality in greater detail, providing visible, audible and/or tactile cues, as a real patient would, to guide the scenario and test the learner. Recent simulation development has been directed towards computerised simulators for surgical and endoscopic procedures in which participants use physical devices to guide instruments on a screen, not unlike a video game. These devices often include sophisticated haptics, providing force feedback to the operator, increasing the realism. It is likely that these technologies will be increasingly incorporated into devices for obstetric simulation. Virtual reality simulation, in which participants are immersed into a computer-generated, interactive environment, may become part of training and assessment in obstetrics. It is important to recognise that while sophisticated, high-fidelity simulators are inherently attractive, they are not always necessary or appropriate for some aspects of training, and when appropriate, they may not yield better results for the learner than simple, low-fidelity trainers.
In obstetrics, hybrids of low- and high-fidelity components consisting of manikins, combined with computers and specialised software, can be used to simulate a routine vaginal delivery, breech delivery, shoulder dystocia or postpartum haemorrhage. Simulated maternal vital signs and a foetal heart rate tracing can be displayed. Other publications have reviewed the historical use of simulators and other currently available commercial products for simulation of obstetric procedures and ultrasound.
Where should simulation for obstetrics take place?
The location used for simulation training will vary based on what is available at each institution. Within a given institution, simulation sessions may be held at different sites based on who the learner is and what you want them to learn. By conducting simulation sessions in clinical areas, such as the labour and delivery suite, operating rooms or prenatal ultrasound unit, learners are better able to become immersed in the scenario and have access to equipment, medications and personnel that they normally have available. An obvious limitation to using clinical areas is that simulation events may conflict with patient care. This may limit the ability to plan in advance a specific time or location; rather, one may need to have a scenario prepared in advance and then run the scenario when the clinical volume permits.
As the sophistication of simulators increases, so also does their size, cost and technical support required for operation. This has prompted many institutions to build simulation centres to consolidate their simulation equipment and experienced personnel. Interaction with technical experts and other specialities at a simulation centre can increase the depth of the simulation experience for the learner. The construction, outfitting and maintenance of a large simulation centre and its equipment are costly, but most institutions should be able to afford a limited set of models for obstetric simulation.
What can be simulated in obstetrics?
The list of obstetric skills and emergencies that can be simulated is extensive. Publications related to simulation in obstetrics typically fall into three categories:
- 1.
description of a simulation model/drill only;
- 2.
description of a model/drill with comparison of learner performance before and after simulation (immediately or at a later date); and
- 3.
description of one or more simulation models/drills incorporated into a multi-faceted and/or multidisciplinary and/or teamwork training programme.
The following is a sampling of recent publications describing the use of simulation for the training and evaluation of performance of key obstetric skills and emergency management techniques. Certainly, these ideas can be expanded upon and additional scenarios created.
Shoulder dystocia
Shoulder dystocia is perhaps the obstetric event most ideally suited for simulation. It occurs infrequently and unpredictably. There is evidence to guide best practice. Ideally, a well-designed simulation programme would teach and test the knowledge and manoeuvres needed to affect the best outcome. Reviews are available describing the development of shoulder dystocia simulations.
Deering and colleagues showed that residents trained with a combination of didactics and simulation improved their performance on subsequent drills. Goffman and associates used a simulated shoulder delivery to test, train and retest a group of resident and attending physicians. Following the training session, residents performed better, especially in the use of the more advanced manoeuvres sometimes required to effect delivery (e.g., rotation, delivery of the posterior arm). The technical performance of the attending physicians was higher at baseline and did not improve following the training. Both groups, however, showed improvement in communication.
A concern with any simulation training is the retention of any skills that were learned. Crofts and colleagues tested providers on a simulated shoulder dystocia delivery and then held a didactic training session. Before training, only 49% were able to deliver the foetus. Participants were tested again at 3 weeks, 6 months and 12 months after the training session, with 82%, 84% and 85%, respectively, able to achieve delivery. Based on these results, they proposed that annual training might be sufficient for most; more frequent practice may be warranted for those initially lacking competence.
Simulated shoulder dystocia deliveries using sophisticated manikins can measure the various forces applied to the region of the brachial plexus. This information can be used to demonstrate to trainees the amount of force they apply during a delivery and guide the adoption of manoeuvres that reduce the amount of strain at this injury-prone area.
Crofts et al. sought to compare the effectiveness of low-fidelity (doll-and-pelvis) to high-fidelity (incorporating force feedback) simulation for training in shoulder dystocia deliveries They found overall improvement in the ability to effect delivery of the foetus, from 43% to 83%. The group trained with force feedback showed greater improvement, with 94% achieving delivery versus 72% in the low-fidelity trained group. The high-fidelity group also had less total applied force. Both models incorporated a patient-actor to interact with the provider during the simulation. Participants were scored on communication before and after with both groups showing similar improvement.
Due to the potential litigation after a shoulder dystocia event, accurate and complete documentation of the event is critical. Deering and colleagues demonstrated that after a simulated shoulder dystocia delivery, documentation by obstetric residents was deficient in some key areas. Similarly, Crofts et al. found inaccuracies and missing data in documentation following a simulated shoulder dystocia delivery by comparing participants’ delivery notes to the video recording of the simulation.
There is limited data on the effect of simulation training of any event and subsequent improvement in patient outcome. Draycott and colleagues compared outcomes of deliveries complicated by shoulder dystocia during a 3-year period before the institution of a shoulder dystocia training programme to a 3-year period after. The delivery and shoulder dystocia rates (2%) were similar during these periods. They found an improvement in choice of manoeuvres used to affect delivery, a decrease in the use of excessive force and a decrease in neonatal injury at birth from 9.3% to 2.3%. Other outcomes at birth were not different. There was a trend towards decreased rates of persistent brachial plexus injury at 6 and 12 months of life after the training was initiated, but this difference was not significant.
Postpartum haemorrhage
Faced with massive obstetric haemorrhage, the obstetric provider must be able to rapidly identify the cause and provide proper treatment. Simulation can be used to teach and test providers on the basic steps of management and key facts, such as selection of medications with appropriate doses.
Deering and colleagues recently described a simulation using a birth manikin with a postpartum haemorrhage kit. The kit includes a uterus that can be inflated or deflated to vary the uterine tone and a reservoir that holds 1 l of blood, which can be delivered from the vagina. Residents were expected to perform an appropriate examination, uterine massage and request at least two appropriate medications at the correct doses. Resident evaluations were discussed and a proper management scheme offered after the simulation. The simulation sessions were graded and they found that many residents were unable to stop the bleeding in the 5-min timeframe allotted or made at least one error. This form of objective testing and immediate feedback offers a richer learning experience compared with the traditional question-and-answer form of teaching.
Preeclampsia/Eclampsia
Eclampsia may be severe, unpredictable and sudden in nature and requires immediate management. Given the relatively standard set of initial management steps (e.g., administration of magnesium sulphate) and actions to avoid (e.g., emergency caesarean delivery for foetal heart rate decelerations during or immediately after a seizure), eclampsia is well suited for simulation. Ellis et al. evaluated the effectiveness of simulation training for eclampsia management by comparing the usefulness of drills and training sessions performed at local hospitals to those performed at a simulation centre. They also evaluated whether additional training in teamwork theory improved management over training in clinical management alone. The simulations at the local hospitals used patient-actors and basic manikins. The simulation centre used an advanced human patient simulator model. Following the initial drills, participants received didactic and hands-on training in the management of eclampsia. Half of each group received additional training in teamwork theory. The simulations were repeated 1–3 weeks later. Video recordings of all simulations were reviewed by graders blinded to whether the drill was performed before or after the training session. Following training, there was an improvement in both the rate of completion of expected tasks as well as the time taken to complete these tasks. There was no difference in the improvement of those trained at the local hospital compared with those trained at the simulation centre. Teamwork training did not increase improvement.
Thompson and colleagues used eclampsia drills to identify deficiencies in management. Patient-actors were used and all levels of staff were included. To simulate the unexpected nature of eclampsia, drills were conducted not only on labour and delivery, but also on inpatient wards and the emergency department. Following a debriefing session, the drill was repeated the same day, in another area of the hospital. The authors identified several areas for improvement and noted that over time management of the simulated patients improved.
Standard manikins can be modified to simulate seizures, potentially adding to the realism of the scenario and increasing the involvement of participants. Motors and actuators with remote controls have been used to cause movement of the manikin in a fashion simulating a seizure typically seen in eclampsia.
Breech vaginal delivery
Planned vaginal breech delivery of a singleton foetus is a rare occurrence in modern obstetrics. However, the unexpected breech delivery will likely be encountered by all obstetricians at some point in their careers. Similar to shoulder dystocia, there are standard recommended manoeuvres, techniques and precautions required to safely deliver a foetus from the breech presentation. Most trainees will not be exposed to an actual term vaginal breech delivery during their training. These factors make breech deliveries another obstetric procedure ideally suited for simulation training.
Deering and colleagues used a simulated vaginal breech delivery to train and evaluate residents. A standard obstetric manikin and articulated foetus were used for the simulation. With no prior notice, participants were asked to deliver the foetus from the breech presentation. Following the initial simulation, a training session including didactics and practice with the model was performed. The participants were tested 2 weeks later with the same model, also without prior notification. They found improvement in performance of the critical manoeuvres and that trainees subjectively performed better and achieved delivery in a safer manner.
The Johns Hopkins residency programme in obstetrics and gynaecology has developed a model for the breech extraction of a second twin. An articulated baby doll is placed in a thin plastic bag to simulate the amniotic sac. This bag is then filled with water, tied closed and placed in a heavy-duty plastic bag. This bag simulates the uterus, with the open end passed through a hemi-pelvis model. The learner is asked to examine the foetus, find and grasp the feet and bring the feet down through the vagina. Then, the thin plastic bag can be broken, simulating rupture of the membranes. The foetus is delivered from the breech presentation. Piper forceps can also be used for delivery of the foetal head to add another element to the simulation scenario.
Operative vaginal delivery
The rate of operative vaginal delivery with forceps or vacuum devices has declined over recent years such that many recent graduating residents are not comfortable in their performance. If this trend continues, the rates will continue to decline and likely contribute to an increase in caesarean delivery rates. Furthermore, there has also been a shift away from forceps to vacuum deliveries. Unfortunately, while vacuum cups may be easier to place than forceps blades, they are associated with increased neonatal risks, especially when used incorrectly. Simulation can be used to teach operative vaginal delivery and allow the trainee to practise the correct placement of forceps blades and vacuum cups as many times as they need without placing a patient or foetus at risk.
Simple models using available forceps, a pelvis model and doll can be used in formal training sessions and for ad hoc demonstration and rehearsal on labour and delivery before an actual delivery. Limited data support that a formal education programme in operative vaginal delivery, including simulation with doll-and-pelvis model, may improve safety for mother and baby. Simulation models of vacuum deliveries have been used to identify the key steps of a successful delivery with subsequent application of this template to the training and evaluation of learners.
One limitation to the bedside teaching of the placement of forceps blades is that the teacher cannot easily assess the movement of the blade during insertion by the learner. Rather, the teacher is limited to assessing general technique and the final location of the blade after placement. Dupuis and colleagues designed a high-fidelity forceps simulator that allows for the tracking of the tip of the forceps blade during insertion. They used their model to compare the performance of junior and senior providers and found that the senior operators inserted blades following a more appropriate trajectory. In follow-up studies, they used this simulation model as part of training sessions for junior physicians and showed an improvement in blade placement.
Leslie and colleagues used a testing unit to measure the isometric strength generated by residents in a standing and seated position. They used visual feedback to train the learners to apply the correct amount of traction, and demonstrated that after practice, the trainees could reproduce the correct forces. In another study following those above, Moreau and colleagues used their model to train and assess learners in the forces required to extract the simulated foetal head.
Needle procedures
Procedures such as amniocentesis and foetal blood sampling require sufficient hand–eye coordination to safely direct a needle through maternal tissue into the desired foetal compartment under ultrasound guidance. The learner must be able to visualise in three dimensions while viewing a two-dimensional ultrasound image. Investigators have described both low- and high-fidelity means of effectively simulating needle procedures.
Several groups have described amniocentesis trainers constructed with sonographic targets suspended in gelatin. These simulators allow operators to practise the hand–eye coordination required to guide a needle to a desired target. Targets can simply be solid objects used to practise guidance of the needle or liquid-filled objects, such as small balloons, which can be punctured for simulation of withdrawal of amniotic fluid. Simple simulations have also been developed to practise other invasive procedures including chorionic villous sampling (CVS) and in utero foetal shunt placement. A hollowed-out guava filled with layers of gelatin can be used to simulate CVS procedures (see Fig. 1 ).
