Postpartum hemorrhage (PPH) is the leading cause of maternal mortality worldwide, contributing to approximately 140,000 deaths per year.1 Because of readily available medical and surgical management, the number of deaths in the United States is low; nonetheless, PPH remains a leading cause of pregnancy-related morbidity and mortality. Early recognition of known risk factors for PPH and the use of third-stage intravenous oxytocin have been shown to significantly lower the incidence of PPH.2–4 Unfortunately, diagnosing clinically significant PPH is notoriously difficult because health care providers typically underestimate blood loss and, in pregnancy, signs of hypovolemia are also delayed. Appropriate management of PPH therefore typically requires an urgent interdisciplinary coordinated effort to safely administer a variety of medications and perform any necessary procedures to stop the patient’s blood loss.
In situ simulation of PPH is a particularly valuable process. First and foremost, it provides a platform for interdisciplinary team practice while maintaining the environmental and system factors present during actual patient care events. Second, it allows the entire interdisciplinary team to practice management of the PPH patient, including diagnosis, medical (nonoperative) management, and resuscitation of the acutely decompensating patient. Finally, the debriefing process allows for the (1) identification of gaps in knowledge and teamwork behaviors and (2) detection of environmental and system deficiencies before an actual adverse event. This article describes the development, content validation, and in situ implementation of a standardized patient-based, interdisciplinary PPH scenario. The training received expedited approval from the University of Washington Institutional Review Board.
This simulation involved nurses (n = 49), obstetrical residents (n = 19), obstetrical attending physicians (n = 9), anesthesiology residents (n = 3), and anesthesiology attending physicians (n = 5). Although teams can vary in size and still successfully complete the scenario, optimal team composition consists of 6 to 8 individuals with expertise from all 3 disciplines/specialties (nursing, obstetrics, and anesthesia.) A total of 83 learners at the University of Washington Medical Center Labor and Delivery (L&D) unit completed the scenario.
Needs Assessment and Learning Objectives
A robust needs assessment is critical to the development of effective training programs.5,6 Thus, scenario development began with a needs assessment for the University of Washington Medical Center L&D unit involving all 3 participating disciplines (nursing, obstetrics, and anesthesia). This process focused on the team training and team-level clinical objectives. Training-specific clinical knowledge and procedural skills were not the focus of the simulation and were therefore not a component of the needs assessment.
The teamwork needs assessment occurred through several mechanisms. First, 2 of the investigators (C.R.A. and T.J.B.) integrally involved in quality assurance for the L&D unit reviewed cases discussed from 2009 to 2011. Events where communication, team skills, or systems-related challenges clearly played a role were identified and further discussed. Recurrent themes were identified as potential training targets. Second, the investigators reviewed teamwork and performance themes that emerged during review of a previous series of in situ simulations that involved interdisciplinary care of a patient requiring an emergent cesarean delivery. Teamwork performance deficits identified by the instructors (C.R.A., L.C.) as well as teamwork issues raised during the simulation debriefing sessions were recorded, and common themes and challenges were noted. Data from these simulations were included in the needs assessment because it was felt that similar higher-level communication and teamwork behaviors would be required to manage both types of decompensating patients (eg, emergent cesarean delivery and PPH). The investigators recognize that case-specific behaviors would likely not transfer between cases and were careful to keep resulting learning points at the level of general obstetrical emergency management. Finally, themes identified through both case review and past simulation review were discussed with the L&D clinical leadership in nursing, obstetrics, and anesthesia to further describe, clarify, and prioritize training objectives. The results of the needs assessment and derived training objectives are described in Table 1.
The clinical care needs assessment was performed by evaluating current standards of practice for patient care during bleeding emergencies.2 The authors conducted an in-depth literature review and evaluation of current local hospital protocols. Critical components were included in the learning objectives (Fig. 1).
Theory and Design
A 20-minute obstetrical bleeding emergency simulation was designed using an event-based approach to training.7,8 An event-based approach to training uses discreet events placed strategically throughout the course of the simulation to ensure that desired behaviors and team interactions are triggered in a realistic manner. Standardized patient, confederate, and clinical events (bleeding, vital sign changes, etc) served as triggers throughout the scenario. The content of the triggers and events was informed by the needs assessment and are listed in Figure 1.
The decision to use a standardized patient–based format was multifactorial. First, because one of the objectives of the simulation centered on communication with the patient and patient-centered care, realistic patient communication cues were essential. Second, because performance of procedural skills and more invasive management were not part of the simulation, adequate fidelity could be obtained using simulation adjuncts (simulated intravenous hook-up, vaginal bleeding with PartoPants, etc). Finally, the use of standardized patients for similar roles has been supported in the literature.9,10
Scenario Content Validity
The scenario clinical events and patient clinical course, clinical indicators and triggers, and expected patient care and teamwork actions were content validated by independent health provider subject matter experts (board-certified obstetricians and L&D nurses with >5 years of experience). The methodology of content validation involved having a sample of subject matter experts rate the extent to which the event content (including sequencing, timing, and requisite responses) appropriately represents the typical resuscitation problem it is designed to simulate (Fig. 2 for example).11 A sample of 17 subject matter experts (9 obstetrical nurses, 8 obstetricians) was used to meet professional standards for content validation (6–8 experts per area of evaluation).12 Content rating for appropriate representativeness used a 5-point Likert scale (ranging from 1, strongly agree, to 5, strongly disagree). The mean rating was 1.84, indicating strong endorsement by the subject matter experts. Indices for overall agreement rwg(j) (0.98) indicated good subject matter expert convergence that scenario events, clinical indicators, and targeted medical actions were representative and met appropriate standards for patient care. This convergence, in combination with the high rating mean, substantiated the content validity of the scenario.
This scenario can be performed in situ or in a simulation center. To optimize our ability to evaluate teamwork and systems factors in our simulation, we conducted simulations on an L&D unit. Phases 1 to 3 were conducted in a standard labor, delivery, and resuscitation room, and the patient was transferred to an operating room for phases 4 to 5 (Fig. 1). All parts of the scenario could be performed in a single location if appropriate for the institution’s practice.
Instructors (2 needed) were board-eligible obstetricians and licensed, certified obstetrical nurses with instructional experience in simulation training and debriefing. Both physician and nursing instructors were present at each debriefing to ensure a balanced, interprofessional discussion of teamwork issues. In addition, each simulation required instructors to assist standardized patients with their verbal prompts by giving reminders through a 1-way radio/headset. This role can be important because teams often use unanticipated strategies that nonclinical standardized patients and nurse confederates may not be prepared to address in a manner that is both realistic and in line with the goals of the simulation. For example, more aggressive management (ie, invasive monitoring or surgically based treatments) would not be possible on a standardized patient. When instructors hear such planning occurring, they can give the confederates realistic responses (“I cannot consent to any procedures without my husband present.”) that allow the scenario to continue without pointing out obvious technical limitations. Having standardized patients recall all timed prompts and respond to such unanticipated learner comments is quite challenging and is facilitated by the ability of an instructor to talk through the response with the standardized patient in real time.
Both the nurse confederate and standardized patient (discussed later) were recruited from the University of Washington Standardized Patient Program. The nurse confederate served to introduce the team to the patient and to help teams provide appropriate care within the simulated environment. She provided assistance when asked but does not help with medical knowledge or clinical care direction. The nurse confederate was also tasked with providing “backup triggers” to ensure the team progresses through the entire scenario as written (Fig. 1). By providing such triggers, the instructors guarantee that the learners have the opportunity to meet all learning objectives. Because the goal was to have the team (not the nurse confederate) complete most tasks, she was introduced as a new, that is, inexperienced, nurse transporting the patient from the emergency department. To ensure consistency and accuracy of role fulfillment, standardized patients received 20 hours of training. Confederate performance was assessed before initiating simulations and throughout the trainings by the investigators (C.R.A., L.C., M.F.F.) who reviewed video recordings and provided focused feedback.
The standardized patient role required a female, age 20 to 40 years. The standardized patient also provided critical triggers and prompts to the team (Fig. 1). As the simulation progressed and the patient’s condition worsened, the standardized patient became more anxious and concerned with her clinical status. Her vital signs were provided using a simulated monitor, and the low-to-high volume vaginal bleeding was simulated using PartoPants.13 The standardized patient controlled the rate of bleeding via a 3-way stopcock.
A minimum of 1 technician was required to set up and run the scenario. Two were required if video recording was used. Overall preparation was 1 hour (2 technicians) for in situ simulations, including camera placement, simulated monitor setup, and standardized patient preparation of PartoPants. PartoPants were selected because we have experience using them, they are inexpensive, and they can be discarded when they become overly stained with simulated blood. Care was required to ensure simulated blood would not stain floors or clothing. A complete list of technical materials and props is provided in Table 2. Although the items chosen were based on a number of factors, it is likely that adequate substitutes are commercially available.
Training sessions lasted for 2 hours in total and involved the following:
- Sign in (5 minutes)
- Introduction to simulation environment and PartoPants (15 minutes)
- Simulation session (20 minutes)
- Break (10 minutes)
- Debrief and feedback to instructors (50 minutes)
- Wrap-up (5 minutes)
At the start of the simulation, all (or a portion) of the participants can be read the following case stem. The decision on who should receive the stem depends on desired learning objectives, the skills of the learners, and the systems-based practices at the training institution.
A 34-year-old G3, P2 female has just been admitted through the emergency department after a precipitous delivery and subsequent postpartum hemorrhage from uterine atony, now controlled after uterotonics (40 U oxytocin IV). Patient’s estimated blood loss is 1.5 L before arrival. She has been given 1 U of packed red blood cells and 2 L of crystalloids. Type and cross match has been performed. (Her laboratory values from the emergency department are available on a chart at the bedside if requested.)
All (or a portion) of the participants can then be directed to the patient’s bedside. At this time, the patient’s chart is available if requested (including laboratory values from her emergency department stay). The patient is able to provide any necessary medical history. The degree of spontaneously provided information can differ depending on the skill set of the learners.
Medical history: None (specifically no bleeding disorders)
Obstetrical history: Normal spontaneous vaginal delivery 2 years before
Patient has had good prenatal care, no complications
Medications: Prenatal vitamins
Family history: Hypertension
Social history: Denies all tobacco, alcohol, drug use
As the team progresses through the case, they may call for help from other members if not initially allowed into the room. This allows for potential changes in leadership, handoffs, and briefs depending on the system in place at the training institution. Updated bedside ultrasound and bimanual examination findings are available to the team as they are requested. The team may also transfer the patient to the operating room if surgical intervention is anticipated. The case ends with stabilization of the patient.
Debriefings were conducted using methodology outlined by Brett-Fleeger et al.14 First, instructors set the tone for psychological safety within the debrief by establishing clear expectations of respect for and between the learners and by reinforcing a standard confidentiality agreement that is part of all simulation-based training at the University of Washington. This was followed by an open question to the participants to elicit general thoughts about the simulation experience. These were noted by one of the instructors acting as a scribe. Next, a short period was spent reviewing the medical knowledge and diagnostic issues raised during the simulation. Specific critical components of the bleeding protocol were discussed, and participants were encouraged to reflect on their performance and any barriers to effectiveness they noted. Instructors also ensured that the process of estimating blood loss and indications for operative management were reviewed.
After this part of the debriefing, instructors led participants toward discussion of teamwork behaviors. Any communication or teamwork issues identified during the first part of the debrief were further explored and deconstructed to see if any themes emerged (see Results). In addition, the instructors used clips from the simulation video to allow participants to observe themselves and reflect specifically on key aspects of (1) leadership and leadership transition, (2) provider-patient communication, (3) role assignment, and (4) the use of team updates to ensure a shared mental model. Before leaving the training, learners completed a training evaluation used by the University of Washington Institute for Simulation and Interprofessional Studies.
TRAINING EVALUATION AND LESSONS LEARNED
This simulation was overall well received. Participants indicated that the simulation was effective (1.64 ± 0.73), and they would like more frequent sessions with different clinical emergencies (1.4 ± 0.7) on a Likert scale rating (1 = strongly agree, 2 = agree, 3 = neutral, 4 = disagree, 5 = strongly disagree). Results from the postsimulation survey, along with participant response rates, are presented in Table 3a. In addition, participant comments were categorized into themes and are summarized in Table 3b. Participants commonly cited teamwork issues (eg, role clarity, resource management, and communication issues) as important learning points. Although the majority of themes present were consistent throughout the simulation sessions, earlier participants noted that the nurse confederate was confusing to them and they were uncertain how to interact with her. As a result, an additional component was added to the presimulation briefing to introduce the participants to the concept of the confederates. In addition, the confederate script was expanded to offer a more realistic role for them in the simulation.
Several challenges to scenario implementation persisted throughout the sessions and are summarized in Table 3c. Overall, requirements for simulation resources and participant time were the greatest obstacles. Even with enthusiastic support from nursing, physician, and administration leadership, blocking off time for adequate preparation and debriefing sessions was challenging. Moreover, the desire to simulate the transfer of the patient to the operating room required that the operating room be reserved for training purposes. Although this was a significant issue, the inclusion of this component to the simulation was particularly valuable because it raised multiple teamwork and communication issues. This simulation could be replicated in its original form or could be scaled down as needed, with the understanding that learning objectives and critical triggers might require adjustment.
1. World Health Organization: World Health Statistics. 2011. Available at: http://www.who.int/whosis/whostat/2011/en/
2. American College of Obstetricians and Gynecologists. ACOG practice bulletin: clinical management guidelines for obstetrician-gynecologists number 76, October 2006: postpartum hemorrhage. Obstet Gynecol
2006; 108 (4): 1039–1047.
3. Rajan PV, Wing DA. Postpartum hemorrhage: evidence-based medical interventions for prevention and treatment. Clin Obstet Gynecol
2010; 53 (1): 165–181.
4. Oyelese Y, Ananth CV. Postpartum hemorrhage: epidemiology, risk factors, and causes. Clin Obstet Gynecol
2010; 53 (1): 147–156.
5. Brooks-Buza H, Fernandez R, Stenger JP. In situ resuscitation simulation to test a pediatric dental clinic emergency response process. Simul Healthc
2008; 3 (4): 254.
6. Goldstein IL, Ford JK. Training in Organizations: Needs Assessment, Development, and Evaluation
4th ed. Belmont, CA: Wadsworth; 2002.
7. Fowlkes J, Dwyer DJ, Oser RL, Salas E. Event-based approach to training (EBAT). Int J Aviat Psychol
1998; 8 (3): 209–221.
8. Rosen MA, Salas E, Silvestri S, Wu TS, Lazzara EH. A measurement tool for simulation-based training in emergency medicine: the simulation module for assessment of resident targeted event responses (SMARTER) approach. Simul Healthc
2008; 3 (3): 170–179.
9. Siassakos D, Draycott T, O’Brien K, Kenyon C, Bartlett C, Fox R. Exploratory randomized controlled trial of hybrid obstetric simulation training for undergraduate students. Simul Healthc
2010; 5 (4): 193–198.
10. Crofts JF, Bartlett C, Ellis D, et al. Patient-actor perception of care: a comparison of obstetric emergency training using manikins and patient-actors. Qual Saf Health Care
2008; 17 (1): 20–24.
11. Haynes SN, Richard DCS, Kubany ES. Content validity in psychological assessment: a functional approach to concepts and methods. Psychol Assess
1995; 7 (3): 238–247.
12. Barrett RS. Content validation form. Publ Personnel Manag
1992; 21: 41–52.
13. Walker D, Cohen S, Monterroso M, Jenny A, Fritz J, Fahey J. PRONTO training for obstetric and neonatal emergencies in Mexico. Int J Gynaecol Obstet
2012; 116 (2): 128–133.
14. Brett-Fleeger M, Rudolph J, Eppich W, et al. Debriefing assessment for simulation in healthcare: development and psychometric properties. Simul Healthc
2012; 7: 288–294.