Job role and stress influence student movement during postpartum haemorrhage simulation: an exploratory study

Abstract

Introduction

Postpartum haemorrhage is the leading cause of maternal death. Healthcare simulations are an educational tool to prepare students for infrequent high-risk emergencies without risking patient safety. Efficiency of movement in the simulation environment is important to minimize the risk of medical error. The purpose of this study was to quantify the movement behaviours of the participants in the simulation and evaluate the relationship between perceived stress and movement.

Methods

N=30 students participated in 10 high-fidelity medical simulations using an adult patient simulator experiencing a postpartum haemorrhage. The participants completed the State-Trait Anxiety Inventory prior to the simulation to measure perceived stress. Physical movement behaviours included walking around the simulation, time spent at bedside, arm movements, movements without purpose, looking at charts/vitals and total movement.

Results

Midwife (MW) students spent significantly more time walking (p=0.004) and looking at charts/vitals (p=<0.001) and significantly less time at bedside (p=<0.001) compared to obstetric (OB) students. The MW students demonstrated significantly more total movements compared to the OB students (p=<0.001). There was a significant, moderate, positive relationship between perceived stress and total movement during the simulation for the MW group (r=0.50, p=0.05). There was a trend for a moderate, positive relationship between perceived stress and total movement during the simulation for the OB group (r=0.46, p=0.10).

Conclusions

Physical movement during a simulation varies by job role and is influenced by perceived stress. Improved understanding of physical movement in the simulation environment can improve feedback, training and environmental set-up.

INTRODUCTION

Postpartum haemorrhage is an obstetric emergency and is the leading cause of maternal death in the world. ¹ Although this medical condition is the leading cause of maternal death, it is still relatively rare, only occurring in <1% of births. ² It is imperative that medical teams are adequately prepared for this emergency, as response time is critical to the patient’s survival. ³ Educational and healthcare simulations provide students with an opportunity to prepare for these types of infrequent, high-risk medical scenarios to improve their technical (required medical skills and competency of such skills) and non-technical (communication and teamwork) skills. ^4–8

Simulations utilizing postpartum haemorrhage as the medical scenario present students with a high-risk and stressful clinical situation in order to challenge and prepare students for this emergency as practicing clinicians. ^9–13 This kind of simulation can be anxiety-provoking, but it can also aide in reducing stress and improving confidence in a low-stakes environment. ^14–17 Perceived stress during a simulation experience can affect performance, learning and confidence. ^18–21 When the participant experiences too much or too little stress, performance may be reduced, along with the quality of learning. ^{11 22 23}

It is also known that physical movement of the participants in a simulation can indicate whether the environment, required materials and training are efficient and optimal. ²⁴ For example, too much unnecessary movement or physical contact with the environment may indicate that the environmental design is not optimal for the task or the participant needs more training to complete the task. ^{25 26} Too little movement may be an indicator that the participants are unsure of their next task step, team communication break downs or poor teamwork, all of which may prevent the patient from receiving needed medical care in a sufficient time frame. ²⁶

Although there is research to suggest that both stress and movement influence performance, there is a current gap in the literature on how stress and movement are related, especially during an educational simulation. In addition, although it is inherent that a specific job role differs between professions and alter movement and behaviour throughout a simulation, little attention has been paid to quantifying these differences to improve design, organization and training during simulations.

Therefore, the purposes of this study were to: (1) objectively quantify movement of student participants during a postpartum haemorrhage simulation utilizing novel video coding, and (2) evaluate the relationship between perceived stress and movement of these participants within the simulation environment.

METHODS

Study design

An observational cohort study design was used to evaluate student movement and stress during a postpartum haemorrhage simulation. This study was performed at the Presage simulation training centre at Lille 2 Université—Droit et Santé.

Participants

The procedures and experiments were approved by the Research Ethics Committee of The Ohio State University (2018E0096). The protocol for this study was considered exempt and consent did not need to be obtained. Thirty midwife (MW) and obstetrician students (N=30; N=20 MW students, N=10 obstetrician students) were included in this study and each participant performed the simulation one time. The sample size was determined based on the number of students enrolled in the simulation as a part of their educational training. As such, all participants were included in this observational study.

The student participants were enrolled in a MW or obstetric programme at Lille 2 Université—Droit et Santé and were of ages 18–28 years old. These students were divided into groups of four. Each student was assigned a role during the simulation. These roles included obstetrics and gynaecology physician (OB) and obstetrics and gynaecology intern (OB intern), MW and MW’s assistant (MW assistant). One person from each of these four designated roles was included in each simulation. Fifth-year obstetric students were assigned as the OB. First-year obstetric students were assigned as the OB intern. MW students were all academically the same year and had the same level of experience, so they were randomly assigned as MW or MW assistant.

Each group, containing an OB, OB intern, MW and MW assistant, participated in a medical simulation using high-fidelity adult patient simulator experiencing a postpartum haemorrhage. Professors were included in the simulation and served as evaluators and anaesthetists during each simulation.

Simulation

The simulation environment was representative of a labour and delivery room in a hospital setting. It was equipped with a high-fidelity female adult simulator (Noelle S550), and ‘real time’ vital signs displayed on a monitor, medical equipment, phones, paper charts and areas for disposal of waste. Two video cameras at different angles captured the entire simulation environment (figure 1). Video was recorded at 60 frames per second. Video cameras were set up prior to the simulation in areas that were out of reach to the participants so that video-recording would not interfere with tasks required during the simulation scenario. Video monitoring of the simulation is a common educational practice.

Figure 1.

Summative total movements of MW and OB groups. The MW group had significantly higher total number of movements. * indicates p =<0.05.

During a postpartum haemorrhage, a specific order of medical treatment needs to occur to ensure the safety of the patient. The order of medical occurrences and treatment needs during the simulation was created based on previous literature. ²⁷ The simulation ran in the following order: (1) the students were introduced to the scenario, high-fidelity mannequin and simulated environment, including the location of any equipment they may need, (2) MW, MW assistant and instructor enter room; MW and MW assistant determine OB intern is needed due to low blood pressure and elevated heart rate, (3) OB intern enters the room and assesses situation; OB intern determines OB is needed due to concern for internal bleeding which is denoted by the continued drop in blood pressure and continued elevation in heart rate, (4) OB enters into the room and assesses situation; OB determines that a postpartum haemorrhage has occurred and surgery is warranted due to the information provided by the previous medical professionals in the room and poor vital signs (low blood pressure and elevated heart rate), (5) anaesthetist is called by the OB, (6) anaesthetist enters the room and receives information from OB and (7) the simulation ends following the decision for surgery.

After the simulation was completed, debriefing occurred using a standardized format. Each participant was given the opportunity to share medical knowledge, discuss teamwork and communication, and express how they would perform differently if they were to do the simulation again. During this time, faculty members provided feedback to the participants.

Measures

The State-Trait Anxiety Inventory (STAI) was administered to students immediately before and after the simulation experience. The STAI is a subjective, 20-item, 4-point scale to measure perceived stress or state anxiety. The internal consistency coefficients have ranged from 0.86 to 0.95. ^28–30 A higher score on this test indicates a higher level of perceived stress. The students also completed a short post-assessment that measured realism, satisfaction and the likelihood of completing the simulation again. Realism was in reference as to how realistic the simulation was, including the simulator and medical scenario. Satisfaction was in reference to the overall satisfaction the participants experienced during the simulation and could include, but was not limited to, satisfaction with the overall experience and learning process. Each of these was measured on a continuous scale from 0 to 10, 0 indicating the lowest score and 10 indicating the highest score. Feedback sheets were completed by the faculty, but the overall performance of the participants was not formally scored. These sheets served as both a checklist of tasks the participants were expected to complete (ie,: check vitals of the patient) and a place to take notes to be provided to the participants during the debriefing session following the simulation.

Video coding

Behavioural coding is a methodology often used in research to objectively quantify specific behaviours using a video-recording. Datavyu, a behavioural coding software, was used for coding. ³¹ Specific movements were defined, identified and measured included walking, time spent at bedside, number of arm movements, fidgeting and time spent looking at the chart and/or vitals. These movements were defined as either: (1) a percentage of time spent doing a movement behaviour while they were in the simulation environment, or (2) a frequency (movement behaviour/min) or the total number of times the behaviour was performed divided by the total time the participant was in the simulation environment. Variables and onset/offset of each variable were operationally defined and are presented in table 1.

Table 1.

Physical movements, units and definitions for onset and offset

Physical movements	Unit	Description
Walking	Percentage: unit defined as a percentage of total time spent walking while the participant is in the simulation environment.	Three consecutive steps in any direction without a >1 s stop. Onset, or start, for this variable was the first frame where the participant begins to move their foot. Offset, or end, for this variable was the first frame following a 1 s stop in movement. Walking could occur in any direction (forwards, backwards or sideways) as long as it met the three consecutive step requirement without a >1 s stop.
Walking bouts	Frequency: unit defined as walking bouts/min.	Total number of walking bouts during the simulation per min.
Time spent at bedside	Percentage: unit defined as a percentage of total time spent at bedside while the participant is in the simulation environment.	Interacting in close and direct proximity of the mannequin-patient who is stationary in the bed. The onset was determined by the first frame the participant is at the bedside providing direct patient care. The participant must be within approximately 2 feet of the bed facing the patient, or they must be physically contacting the patient. The offset for this variable is whenever the participant is no longer within 2 feet of the patient, no longer is directly facing the patient, or no longer in physical contact with the patient.
Arm movements	Frequency: unit defined as arm movements/min.	Upper extremity movements that were not a part of a functional task (ie, reaching for pen is a functional task). Onset of this variable began at the first frame the participant starts moving their upper extremities. The offset for this variable was defined as the first frame the participant is no longer completing the upper extremity movement, or the participant begins a functional task. Arm movement number was the total number of arm movements the participant completed during the simulation.
Fidgeting movements	Frequency: unit defined as fidgets/min.	An abrupt change in direction, or weight shift, without a goal-oriented task associated with the movement. The onset for this variable is the beginning of a non-goal-oriented weight shift or abrupt change in direction (ie, the participant begins walking in a direction, but realizes the equipment they need is behind them, leading to an abrupt change in direction). The offset for this variable is whenever the participant returns their own body to midline. Fidgeting number was the total number of times the participant ‘fidgeted’ during the simulation.
Time spent looking at the charts or vitals	Percentage: unit defined as a percentage of total time spent looking at charts or vitals while the participant is in the simulation environment.	When the participant was actively looking at the charts or vitals, which meant the head direction, eye gaze or entire body is facing areas in which charts and vitals are located. Offset for this variable occurs when the participant is no longer looking at the charts or vitals. Looking at charts and vitals number was the total number of time the participant looked at charts or vitals during the simulation.
Look number	Frequency: unit defined as looks/min.	Total number of times the participant looked at either the chart or vitals throughout the simulation.
Total movement	Frequency: unit defined as movements/min.	Summative variable was the total of the frequency variables. This included frequencies of total number of starts/stops, looking at charts/vitals, fidget movements and arm movements. This variable is used to represent the total number of movements, or frequency, of all variables that were coded.

Movements with the unit indicated as ‘Percentage’ are described as the percentage of time spent doing a movement behaviour divided by the total time the participant was in the simulation. Movements with the unit indicated as ‘Frequency’ are the total number of times the participant completed that movement behaviour divided by the total time the participant was in the simulation.

Statistical analysis

Descriptive statistics, including means, SD and frequencies, were used to describe important simulation time points, movement variables, stress, realism, satisfaction and the likelihood of the student participating in the simulation. Non-parametric statistics were used to evaluate differences between groups due to the small sample size of the groups (N=20 for MW and N=10 for OB), lack of normality of the data sets, and study design. Mann–Whitney U-test was used to evaluate differences between group behavioural movements and perceived stress. Spearman correlations were used to describe the relationships between perceived stress, measured using the STAI, and total movement. A p value of ≤0.05 was considered significant and a p value of ≤1.0 was considered a trend. All statistical analyses were computed using IBM SPSS version 23.0.

RESULTS

The average duration of the simulation was 19.8±4.7 min. The average length of time until the OB intern entered the room was 5.2±3.2 min, until OB entered the room was 9.1±3.2 min and time until anaesthetist entered was 14.6±3.9 min.

The results for the five specific movement variables, including walking, time spent at bedside, number of arm movements, fidgeting and time spent looking at the chart/vitals, are listed below:

Time spent walking was significantly higher in the MW group compared to the OB group (MW=15.8%±5.8%; OB=9.5%±6.3%, p=0.004). There was no difference found between MW and MW assistant group (MW=13.8%±5.0%; MW assistant=8.7%±4.7% p=0.19) or OB group and OB intern group (OB=10.3%±7.7%; OB intern=5.3%±1.9%, p=0.73).

Walking bouts were significantly higher in the MW group compared to the OB group (MW=2.3±0.6 walking bouts/min; OB=1.3±0.7 walking bouts/min, p=<.001). There was no difference found between MW group and MW assistant group (MW=2.0±0.5 walking bouts/min; MW assistant=2.5±0.5, p=0.06 walking bouts/min) or OB group and OB intern group (OB=1.5±0.9 walking bouts/min; OB intern=1.0±0.4 walking bouts/min, p=0.44).

Time spent at bedside was significantly higher in the OB group compared to the MW group (MW=40.1%±19.3%; OB=72.6%±19.7%, p=<0.001). There was no difference found between MW group and MW assistant group (MW=43.1%±19.5%; MW assistant=36.9%±19.3%, p=0.52) or OB group and OB intern group (OB=72.0%±20.1%; OB intern=73.1%±20.3%, p=0.97).

Arm movements were not different between MW group and OB group (MW=0.7±0.4; OB=0.6±0.6, p=0.33), between MW group and MW assistant group (MW=0.9±0.7; MW assistant=0.6±0.4, p=0.31), or between OB group and OB intern group (OB=0.9±0.7; OB intern=0.4±0.3, p=0.14)

Fidgeting movement was significantly higher in the MW group compared to the OB group (MW=0.4±0.3 fidgets per min; OB=0.2±0.2 fidgets per min, p=0.01). There was no difference found between MW group and MW assistant group (MW=0.3±0.2 fidgets per min; MW assistant=0.5±0.4 fidgets per min, p=0.08) or OB group and OB intern group (OB=0.2±0.2 fidgets per min; OB intern=0.2±0.1 fidgets per min, p=0.97).

Time spent looking at the charts or vitals was significantly higher in the MW group compared to the OB group (MW=14.6%±10.9%; OB=2.1%±2.9%, p=<.001). There was no difference between the MW group and MW assistant group (MW=13.8%±0.9%; MW assistant=15.3%±0.5%, p=0.35) or between the OB group and OB intern group (OB=2.7%±3.7%; OB intern=1.6%±1.8%, p=0.65).

Look number was significantly higher in the MW group compared to the OB group (MW=1.0±0.6 looks per min; OB=0.3±0.4 looks per min, p=<0.001). There was no difference found between the MW group and the MW assistant group (MW=0.7±0.4 looks per min; MW assistant=1.3±0.6 looks per min, p=0.06) or the OB group and OB intern group (OB=0.4±0.5 looks per min; OB intern=0.2±0.2 looks per min, p=0.84).

Total movement was significantly higher in the MW group compared to the OB group (MW=5.1±2.4 movements per min; OB=2.6±2.0 movements per min, p=<0.001). OB group also had significantly more total movements than the OB intern group (OB=3.6±2.4 movements per min; OB intern=1.6±0.5 movements per min, p=0.02). There was no difference found between MW group and MW assistant group (MW=5.1±3.1 movements per min; MW assistant=5.0±1.2 movements per min, p=0.96) (figures 1–3).

Figure 2.

Summative total movements of MW and MW assistant groups. Summative total movements. No difference found between groups.

Figure 3.

Summative total movements of OB and OB intern groups. The OB group had significantly higher total number of movements. * indicates p =<0.05.

Analysis of perceived stress included only those participants who had STAI data (N=16 MW group and N=10 OB group). There was not a statistically significant difference between perceived stress prior to the simulation (MW=53.2±7.8; OB=49.6±7.3, p=0.15). The MW and OB group relationships between perceived stress and TM were analysed separately, as there were significant differences found between these groups for TM. There was a significant, moderate, positive relationship between perceived and total movement during the simulation for the MW group (r=0.50, p=0.05). There was a moderate, positive relationship between perceived and total movement during the simulation for the OB group (r=0.46, p=0.10). These results both indicate that higher perceived stress prior to the simulation was related to increased total movement during the simulation (figures 4 and 5).

Figure 4.

There was a significant, moderate, positive correlation between total movement and stress was found for the MW group.

Figure 5.

There was a moderate, positive correlation between total movement and stress was found for the OB group.

Scores for realism, satisfaction and likelihood of completing the simulation again were recorded on a visual analogue scale and rated from 0 to 10, with 10 being the highest score. Both MW and OB students completed these scales as a portion of their post-assessment. The average realism score was 7.2±2.1, the average score for satisfaction was 8.8±1.0 and the average likelihood of the student choosing to participate in the simulation again was 9.2±1.0.

DISCUSSION

The results from this study suggest that (1) video analysis of simulations can be utilized to provide objective information on student movement, (2) movement around the simulation environment is related to specific job roles when measured objectively, (3) perceived stress prior to a simulation is related to movement throughout a simulation, regardless of job role and (4) student acceptability of simulation training is high. The results from this study demonstrate that video-recording and analysis can provide a robust amount of information from a medical simulation for research purposes. ³² Physical movements were chosen specifically for this healthcare simulation, but, importantly, any behaviour could be objectively quantified using this technique.

MW students spent significantly more time walking around the simulation environment, looking at charts and vitals, and had significantly more total movements than the obstetrician students. It is of note that all these variables were normalized to the total amount of time the student was in the room. This highlights the important role midwives have as early identifiers of possible medical emergencies and then subsequently the supporting role once the obstetrician is present. In contrast, obstetrician students spent significantly more time at the bedside with the patient, where they were directly providing patient care. Teamwork and communication between these roles are needed to ensure efficiency and patient safety in postpartum haemorrhage emergencies.

The movement behaviours described from this study support previous literature that medical rooms could be set up to support the unique roles of the medical professionals working in them. ^{25 26} Midwives moving around the room most often suggests they maybe most at-risk for inefficiencies that would delay patient care during a medical emergency. It may also suggest they need additional support to complete the necessary tasks. Medical rooms could be setup to reduce movement to improve timely care. In addition, physical environments can be set up to account for movement patterns exhibited by the participants to decrease the risk for negative encounters within the physical space, such as bumping into a tray table or having to walk back and forth to get supplies. ³³

In addition, the results suggest that higher levels of perceived stress, regardless of job role, lead to increased physical movement in the simulation environment. Student participants with higher perceived stress prior to the simulation moved more throughout the simulation environment compared to participants with lower perceived stress. This supports previous literature that excessive movement in the simulation environment can lead to delayed care and inefficiencies in medical treatment, which may negatively impact patient outcome. ^{25 26} Additional practice in a simulated, low-stakes environment may be used to reduce stress and optimize movement for participants through exposure to high-stakes medical scenarios, training and targeted feedback. ^{15 16 34–36}

Finally, students generally have a high acceptability for the use of simulation training and video-recording with an average score of 9.2 out of 10 to complete the simulation training again. Students were also highly satisfied with the simulation, rating satisfaction at 8.8 out of 10. These findings support previous literature that simulation training for postpartum haemorrhage is a positive experience for MW and obstetric students. ⁷

Limitations and future directions

Limitations in this study include: (1) the participants in this study were enrolled in programmes at the same university, which influence their training. Their training has implications on how they move about the simulation environment and how they have been taught to deal with stressors. (2) It is unknown whether the participants had ever participated in an educational simulation prior to this study. Exposure to previous educational simulations may have impacted their perceived stress. (3) Since all subjects were videotaped for this study, we do not know if video-recording of the simulation influenced movement or stress exhibited by the participants. It is common for simulations to be recorded and for cameras to be placed around the simulation environment for preceptors to view on a monitor. (4) Unequal and small sample sizes between the groups limit external validity for the results of this study, and not all participants in the study completed all formal measures (ie, STAI). Finally, (5) student performance was not evaluated in this simulation due to the formative pedagogy (low-stakes approach) utilized for training as is common in simulated medical education.

Future work could include more healthcare professionals in both education and simulated clinical healthcare settings. This would ensure that more medical professions are represented. In addition, repeated simulation experiences utilizing the same tools may help further explain movement and perceived stress. We would hypothesize, based on previous literature, that more exposure to simulations would result in less perceived stress, and therefore, less movement around the simulation environment. ^{15 16 34–36} We also suggest it would be important to consider student performance in future studies and evaluate the relationship between perceived stress, movement and performance.

What is already known.

• Medical professionals inherently move differently based on their job role.

• Too much or too little stress can result in poor performance during a medical simulation.

• Perceived stress may be related to performance, learning and confidence.

What this study adds.

• Midwife students move around the simulation environment significantly more than obstetric students.

• Perceived stress is related to movement.

• Higher perceived stress prior to an educational simulation results in more movement during the simulation which may impact efficiency and patient safety.