Use of Cognitive Task Analysis to Understand Decision-making for Management of Blunt Abdominal Trauma in Children

Study objective To use cognitive task analysis (CTA) to elicit experts’ knowledge and outline their clinical decision-making process for the management of pediatric blunt abdominal injury. Methods This was a mixed-methods study involving in-depth interviews and a focus group with local experts followed by a review survey with a panel of experts from numerous pediatric trauma centers across the United States. All experts specialized in pediatric emergency medicine and pediatric trauma surgery. Results Common themes that emerged during seven in-depth interviews with local experts included: clinical management, clinical reasoning, situational awareness, potential errors/novice traps, knowledge and skills, communication, and quality indicators. A total of 17 external experts responded to the survey for a response rate of 77%. External experts indicated that the information outlined in the CTA was complete and accurate, and provided valuable insights into the discrepancies that were unresolved during the focus group with local experts. They indicated agreement with potential errors/novice traps reported by local experts. Specifically, they indicated that the failure in coordinating team activities, maintaining the big picture, and performing a thorough physical examination posed serious threats to novices when managing a child with blunt abdominal trauma. Conclusion CTA may be applied to pediatric management of blunt abdominal injury to identify critical steps and potential errors to serve as a framework for education in pediatric emergency medicine and pediatric trauma surgery. Future studies may apply CTA to other types of traumatic injuries and/or involve an interdisciplinary approach.


Introduction Background
Injury is the leading cause of death in children from ages 1 to 18 years [1]. The severely injured child demands prompt medical intervention for an optimal outcome. Since 80% of children are first cared for by emergency medical services (EMS) and in community hospitals, these healthcare systems must be prepared [2]. Caring for the injured child presents unique challenges for healthcare providers. For example, to detect derangements providers must be familiar with normal values for vital sign ranges, which change with age and are significantly different from adults. Unlike adults, the early recognition of shock in children can be very challenging. An increased heart rate may be the only sign of shock. However, there are numerous other causes of increased heart rate in children, such as pain or anxiety, which confound decision-making. Additionally, treatments and interventions require child-specific decisions be made, in terms of equipment, medication dosage, and definitive care. To remedy this situation, practitioners need an avenue to learn about the management of pediatric trauma patients, specifically to improve their decision-making skills and to reduce the risk of potential errors. Training focused on modeling expert performance can help enhance a practitioner's ability to manage emergent trauma cases. For this complex task, simply observing expert behaviors is not enough. It is crucial to understand the cognitive decisions that underlie expert actions [3]. Decision-making for management of traumatic patients is second-nature for experts, and it may be challenging for them to provide a detailed account of their thought process.

Importance
Cognitive task analysis (CTA) is a method to elicit expert knowledge by dissecting a task or skill into critical steps and by identifying the associated thought processes. This method helps uncover the cognitive processes underlying decision-making that are not accessible to direct observation [4][5][6][7]. In other words, CTA is a way to understand how an expert thinks by identifying cognitive activities needed to complete a task. The insights into experts' decisionmaking processes may provide a useful framework for teaching and learning. CTA has been used to gain a deeper understanding of the thought processes applied by surgical subspecialists while performing specific procedures such as endoscopic retrograde cholangiopancreatography, vaginal hysterectomy, central venous placement, and laparoscopic procedures [5,6,[8][9][10]. Data from these studies led to the development of procedural checklists for use in training and assessment.

Goals of this investigation
Acute trauma management presents as an ideal clinical scenario to apply CTA to assist practitioners in making life-altering decisions regarding management of the traumatically injured child. We sought to use the CTA methodology to gain a more thorough and detailed understanding of the decision-making process for managing acute pediatric blunt abdominal trauma, the most common unrecognized fatal injury and the third most common cause of pediatric trauma mortality [11].

Study design and setting
This study obtained exempt status from the local institutional review board. The design and reporting of this study were based on the consolidated criteria for reporting qualitative research (COREQ) guidelines, a 32-item checklist to report key characteristics of qualitative research studies [12]. This was a mixed-methods study conducted in two phases. The first phase was a qualitative study involving in-depth interviews and a focus group with local experts, which aimed to develop a deeper understanding of the expert clinical decision-making process for managing pediatric abdominal trauma patients. In the second phase, we employed a review survey with a panel of experts from numerous pediatric trauma centers across the United States to triangulate the results that emerged in the qualitative study.

Phase I -Qualitative study/selection of participants
Semi-structured interviews were conducted with local experts at a single site, a pediatric regional trauma center in the Midwest with an annual census of 75,000. A total of seven experts (four pediatric emergency medicine (PEM) attending physicians, two pediatric trauma surgeons and one pediatric trauma surgery fellow with prior experience as a surgeon) from our institution were recruited to participate in this study. Experts were defined as board-certified PEM attendings or board-certified surgeons with >5 years of experience in management of pediatric blunt abdominal trauma, experience as trauma team leaders, involvement with graduate medical education as fellowship directors or as preceptors for trainees, and/or involvement as active members of trauma committees at local or national levels.
Data were collected during a private audio-recorded interview with each expert. Interview length was a maximum of 1.5 hours. Experts were interviewed separately to avoid premature consensus building. Standardized interviews were developed based on evidence-based literature including Advanced Trauma Life Support (ATLS) course [13] and a published clinical pathway for managing pediatric patients with blunt abdominal trauma [11]. Interviews were semi-structured to provide organization for the interviewer and allow the experts flexibility to explore their clinical decision-making processes beyond the standard interview protocol when necessary. Interview protocol included two clinical vignettes to gain insight into experts' internal thought processes when managing a child with blunt abdominal trauma (refer to Table  1). Medical information related to each case was revealed gradually to simulate the dynamic nature of real-life information gathering and to expose experts' evolving decision-making processes. The interviewers asked the experts probing questions regarding the thought processes behind their clinical decisions (including the timing, and order or sequence of management steps) as they explained how they would manage each of the pediatric blunt abdominal trauma cases. Each expert was prompted with open-ended interview questions to explore critical decision points, potential errors/novice traps, helpful sensory cues, necessary knowledge and skills, and quality indicators that informed their management of pediatric blunt abdominal trauma. Key open-ended questions used during the interviews are listed in Table 2.

Case 1
In the ED, you receive a phone call "We have an 11-year-old female who was in a motor vehicle accident, she was restrained and is complaining of abdominal pain. Vital signs: heart rate 130 beats/min, blood pressure 105/70 mmHg, respiratory rate 20 breaths/min, and oxygen saturation 98% on room air. Estimated time of arrival: 5 min".
EMS arrives and report that she was restrained in the rear driver's-side seat of the vehicle with a lap and shoulder belt when the vehicle was struck at high speed on the driver's side. On arrival to the ED, she is awake and alert, immobilized with a cervical collar. Her vital signs are: temperature 37.5, vital signs as reported previously. She is able to maintain her airway and has clear and equal breath sounds without increased work of breathing. She has strong distal pulses. She complains of abdominal pain. Her abdomen is soft and non-distended but she has localized tenderness in the left upper quadrant. There are no bruises or abrasions noted on the abdomen.

Case 2
As you are managing your case, the charge nurse notifies you that "the colliding car has a 12-year-old patient who will arrive in 5 minutes. He was restrained in the rear passenger's side seat of the vehicle with a lap and shoulder belt when the vehicle was struck at high speed on the driver's side. He has a cervical collar in place" [Note: you are the only attending in the ED]. His vital signs are: temperature 37.4, heart rate 135 beats/min, blood pressure 80/50 mmHg, respiratory rate 20 breaths per minute and oxygen saturation 99% on room air. On arrival to the ED, vitals as mentioned previously, he is awake and alert, immobilized with a cervical collar. His airway is patent and he has clear and equal breath sounds bilaterally without increased work of breathing. He has weak distal pulses and his capillary refill is 4 seconds. He complains of abdominal pain and left shoulder pain. On examination, his abdomen is soft and non-distended but he has localized tenderness in the left upper quadrant. He has no bruises or abrasions noted on the abdomen. His genitourinary examination is unremarkable.  The interviews took place in the hospital setting in a private conference room, during the workday, and without interruptions. Each participant provided written consent for this voluntary activity. The principal investigator (T.A.), a second-year fellow in PEM, conducted the interviews after receiving training in cognitive task analysis by a professionally trained instructional designer/educational researcher (S.T.) and co-leading a pilot interview with S.T. T.A. was present at all interviews, S.T. co-led six interviews and C.K. co-led one interview. Audio recordings of the interviews were transcribed by a medical transcriber and reviewed by the investigators for accuracy. The audio recordings contained no participant personal identifiers, such as the name or title of the participant. Audio recordings were deleted upon completion of the study. No repeat interviews were required.

Analysis
The written transcriptions were analyzed using a hybrid thematic analysis approach [14]. This hybrid method, combining deductive and inductive thematic analysis, allowed the researchers to use the existing evidence base on medical decision-making/clinical-problem solving and the published clinical pathway for managing pediatric patients with blunt abdominal trauma as a framework to create a coding scheme [5,[15][16][17]. This hybrid methodology was flexible to extend the coding scheme beyond the initial framework based on the data to account for more taskspecific and covert expert cognitive reasoning processes underlying the observable skills that have not been addressed in the literature previously [16]. See Table 3 for the coding scheme.

Situation Awareness
Gather initial information, use sensory cues, take anticipatory steps (prepare team/equipment), assess  FAST: Focused assessment with sonography for trauma The transcribed interviews were analyzed using the qualitative data analysis software, NVivo for Mac (QSR International Pty Ltd., Version 11. 3.2, 2016, Melbourne, Australia). Two researchers (T.A. and S.T.) independently coded the transcribed interviews using the above-mentioned coding scheme. The data led to the development of sub-themes. Identified themes were compared and coding disagreements were resolved by an iterative process with consensus. A third researcher (C.K.) resolved any remaining minor disagreements. Data obtained from the coded transcriptions were combined into the final set of themes and discrepancies (observed among experts) representing the experts' cognitive reasoning processes for managing pediatric abdominal trauma cases.
The data from the interviews were reviewed by the participating internal experts during a 90minute focus group session with the aim of reviewing the collected data and discussing discrepancies with the experts to understand their rationale for decision-making. The focus group included five of the seven internal experts. The session was facilitated by two researchers (T.A. and C.K.). It was audio-recorded in the hospital setting. This served as a member check to ensure the accuracy of qualitative data.

Phase II -External expert survey
Researchers suggest that triangulation helps strengthen the reliability and validity of a qualitative study [18][19][20]. Certain policies, regulations, and practices specific to the local study site could have influenced the clinical decision-making processes that emerged in the qualitative phase. Therefore, as the last iteration of the CTA, external experts in pediatric trauma across the United States were recruited to review the cognitive task analysis report in the form of a survey.

Selection of participants
A total of 11 PEM attendings and 11 pediatric trauma surgeons from other facilities were identified through peer nomination. They all met the above inclusion criteria for experts and were invited to participate in the external expert review. The participating experts represented diverse geographical locations across the United States. Experts were contacted via email and each expert received one reminder to complete the survey. An online expert review survey was created using Qualtrics software (http://www.qualtrics.com, Qualtrics, Provo, UT). The survey items were based on the key findings and discrepancies related to management of pediatric blunt abdominal trauma highlighted in the first phase of the study (see Table 4 for key questions in the survey for external experts). External experts did not have access to identifiable information. Their role was to review the key findings from the interviews and focus group and to help resolve any discrepancies in the form of a survey. The anonymity of external experts was maintained.

Key Questions
Timing for completion of primary survey

Analysis
The external expert survey data were exported to Microsoft Excel (Microsoft, Redmond, WA) for analyses. Categorical data were presented in frequencies (and percentages), and numerical data were provided as means (standard deviations and 95% confidence intervals). For the ordinal data related to time intervals of management steps, we converted each time interval to a numerical value (i.e., > 0 to < 1 minute = 1; > 1 to < 5 minutes = 2, etc.) and calculated the mean, standard deviation, median, and mode based on this conversion. Examination of all three central tendency measures helped determine the corresponding time interval by which each of the management steps should be completed.

Phase I -Qualitative data analysis results
Interviews were completed with seven experts. The median interview time was 56 minutes 48 seconds (range 32:57-83:10) with a median word count of 7,917 (range 5,559-14,665) and a total sum of 31,686 words. Common themes were described in the coding scheme and the frequencies and relative frequencies from each sub-theme are listed in Tables 5-11. Table 12 shows representative quotations pertinent to each theme.

Clinical Management
Experts shared their decision-making steps to manage a child with blunt abdominal injury. They stated the importance of triaging patients, even before the patient arrives via EMS, by listening for sensory cues, such as, listening to the tone of the EMS provider and the sound of sirens, and interpreting vital signs based on age when deciding to activate a trauma. Experts emphasized the importance of forming goals, such as performing a quick yet thorough primary examination with emphasis on managing the patient's airway and hemodynamic stability followed by conducting a systematic secondary survey. An expert emphasized the importance of teaching critical steps to manage traumatically injured children, "I teach every resident the same thing -a head to toe evaluation, whether they come in with abdominal pain, leg pain etc…the key in all traumas in the ER is your initial overall survey" [sub-theme: perform primary survey; perform secondary survey]. Experts also highlighted the importance of giving crystalloid fluids as needed and re-evaluating the patient after each intervention. Moreover, experts discussed the challenges associated with co-managing trauma patients, such as mixing up orders, labs, or medications. Experts noted the importance of re-evaluating multiple patients without minimizing the patient that initially did not present as sick as the other patient. Experts discussed the benefits of training and the importance of establishing a shared standard vocabulary when evaluating the patient. There were discrepancies among experts about the use of focused assessment with sonography for trauma (FAST) examinations and use of pelvic X-rays.

Clinical Reasoning
Experts offered a detailed account of their thought processes for ordering tests and imaging studies, hypotheses, as well as, reviewing and interpreting data.

Situational Awareness
This refers to identifying, processing and comprehending information to anticipate subsequent steps. It is essential for a trauma team leader to have situational awareness in order to prepare the critically injured child and the team for the next steps in management. The most common sub-themes that emerged for situational awareness included trauma activation/consult, gathering initial information and anticipatory steps. Additional sub-themes included: triaging patients, assessing the situation, prioritizing tasks, using sensory cues, maintaining the big picture, and remaining vigilant. An expert stated that "it's best if you can teach trainees to be hands off at a 30,000 foot view -looking at everything that's happening" to consolidate information and maintain the big picture in order to make key decisions in management. Additionally, an expert emphasized the importance of prioritizing tasks while managing a critically injured child. The expert questioned "what's going to kill the patient and what's going to save the patient?…A shoulder injury is not going to kill the patient".

Potential Errors/Novice Traps
Experts thought that tunnel vision, incomplete physical examination, and failure to maintain big picture could pose as some of the most common potential errors in the management of a pediatric trauma patient. An expert stated that "with novices it's hard to ignore the obvious -if blood and guts are hanging out, they focus on that" [sub-theme: tunnel vision, fixation and anchoring]. Another expert described a novice trap as "trying to give multiple diagnoses…it's easy to say, "Okay, it's probably a clavicle injury and a spleen" and not thinking about, how can abdominal pain make your shoulder hurt?" [sub-theme: tunnel vision, fixation and anchoring]. Another expert stated that "90% of where people will mess up, is…going out of order, not being systematic…head to toe... when the person doing the evaluation goes off of that standard, things get lost, they don't know where they left off, they don't know what they've examined, they don't know if they have looked at the chest or the back or they get confused. And once the leader gets confused, then everyone's confused" [subtheme: incomplete or non-systematic physical examination]. Another potential error is delayed recognition of abnormal vital signs and shock. An expert specified, "one of the biggest things we notice in trauma reviews is a delayed recognition of shock" and "tachycardia is one of the Experts agreed that speed is crucial in the management of a traumatically injured child. Time to complete the primary and secondary survey, time to get the patient to the computed tomography (CT) scanner and time to disposition are priorities. Other factors identified included systems-based care, optimizing the number of team members, performing standardized care and using written orders instead of verbal commands.

Focus Group with Local Experts
Experts reviewed the information collected during the interviews and confirmed their perspectives were captured accurately. Discrepancies among experts were discussed and mostly resolved during the focus group. Unsettled discrepancies included the timing of primary and secondary survey, the timing of chest X-ray (CXR), the utility of pelvis X-ray and use of FAST examinations. Internal experts at the focus group stated the primary survey should take between 30 seconds to 5 minutes and the secondary survey should be completed in 10 minutes. Some experts felt that the CXR should occur with the primary survey, whereas others thought it should take place during the secondary survey. Some experts stated that pelvic Xrays may not be useful, particularly if the patient gets a CT scan of the abdomen and pelvis. Regarding FAST exams, some experts questioned its utility if the patient will ultimately get a CT abdomen/pelvis or require transfer to the operating room. Some experts stated that FAST exams are helpful to evaluate for cardiac tamponade; however, they may not help with patient disposition such as surgical intervention.

Phase II -External expert review survey results
A total of 17 experts (out of 22) across the United States responded to the survey for a response rate of 77%. The external experts indicated that the information outlined in the first phase of the study was representative of the expert clinical decision-making process for managing pediatric blunt abdominal trauma cases. They also provided valuable insights into the discrepancies that were not resolved during the focus group with local experts.
The external experts recommended that the pre-arrival events should occur in the sequence listed in Table 13. Some of these steps could occur concurrently and the order could be adjusted for other cases. All experts agreed that these events should take place before the child arrives at the emergency department.   Table 14 for a complete list of management steps, and corresponding time intervals).  Based on the external experts' responses, laboratory tests for a hemodynamically unstable patient (refer to Table 1, Case 2) should include: complete blood count (n = 17, 100%); liver function tests (n = 13, 76%); lipase (n = 13, 76%); type and screen (n = 10, 59%); basic metabolic profile (n = 10, 59%); urine dip (n = 9, 53%); urinalysis (n = 9, 53%); and type and cross (n = 9, 53%). Additional labs suggested by external experts included a coagulation panel, blood gas, and lactic acid.
There were discrepancies among our internal experts regarding imaging studies, specifically, the timing of CXR, use of FAST exams and use of pelvic X-rays. Thirteen (76%) of the experts agreed that the CXR should be obtained after the primary survey, mostly stating that it should be obtained during the secondary survey after rolling the patient. Experts emphasized that the CXR should not delay resuscitation. Regarding FAST examinations, most external experts, 12 (71%), stated that ultrasound is useful to diagnose cardiac tamponade. Over half of the experts, nine (53%), stated that FAST examinations are helpful in detecting free fluid; however, CT abdomen/pelvis reveals more information. Eight experts stated that FAST exams are helpful for evaluating whether hemodynamically unstable patients should go to the operating room. Nearly one-third of the external experts, five (31.25%), stated that there are no contraindications to performing a FAST examination. A few experts, three (29%), thought that FAST examinations are not helpful in the disposition of a patient. Experts commented that FAST examinations contribute a data point to be combined with other factors. An expert mentioned that in hemodynamically unstable patients, FAST exams are helpful in determining where to start the operation: abdomen, heart, head or chest. Another expert commented that FAST exams are not useful for stable patients and another expert stated that there is no definite information on pediatric FAST examinations yet. Our data suggest that further studies on the utility of pediatric FAST examinations are needed.
Regarding the pelvic X-ray for the second patient, the majority of experts, 14 (82%), said they would consider obtaining a pelvis X-ray if there was a high-speed mechanism of injury in a patient who is hypotensive and has abdominal pain, to rule out a major source of bleeding, if the patient is unstable for CT due to an unstable pelvic fracture. Other experts commented that they would not obtain a pelvic X-ray unless the pelvis was unstable on exam or if they were obtaining a CT abdomen/pelvis.
As to volume resuscitation, 15 (88%) experts agreed that the patient in case 2 needed volume resuscitation. They rationalized that the patient was in hypovolemic shock (tachycardic with delayed capillary refill and weak pulses). The majority of experts, 11 (65%), agreed that the fluid volume resuscitation should be normal saline or lactated ringer's 20 cc/kg. Experts commented that they would follow this with blood. One expert stated that it would be ideal to start with blood, but the delay in obtaining blood from the blood bank obliges the physician to start with normal saline or lactated ringer's solution. Another expert suggested that crystalloid may contribute to coagulopathy. Experts commented that they would activate the massive transfusion protocol.
The potential errors/novice traps reported by local experts were validated by the external experts. Overall, external experts indicated that the following errors could occur commonly/very often: not coordinating team activities, n = 11 (65%); not maintaining the big picture, n = 9 (53%); and incomplete physical examination, n = 9 (53%). Additionally, responses indicated that omission bias, delayed recognition of abnormal vital signs, tunnel vision, lack of prioritizing tasks and missing sensory cues could occur occasionally. External experts suggested additional potential errors: lack of identified team leader, poor crowd control, failure to revise, failure to summarize, failure to perform serial examinations with emphasis on assessment from the initial encounter, over-ordering/relying on tests, over-testing due to liability concerns, poor consultant to consultant interactions, delays in definitive care and lack of ownership. By exploring the cognitive activity of pediatric trauma experts across the United States, we were able to condense their knowledge and experience into a tool for learners (see Table 15 for common pitfalls and Tables 16-19 for comprehensive tools), which we will further explore in the discussion.

Pre-arrival
Missing sensory cues Pay attention to the tone of voice or urgency of EMS (e.g., sirens in the background) Lack of prioritizing tasks Determine if the child is sick or not sick based on the vital signs, the mechanism and obvious injuries

Discussion
Proper pediatric trauma stabilization requires provider knowledge, skills, and behaviors that are child-centered. Providers must be prepared to exercise sound clinical judgment and conduct a well-organized initial assessment of the injured child, recognize the patient's acuity level, and make decisions in a deliberate stepwise fashion to succeed. Community hospital providers particularly experience difficulty maintaining this level of proficiency as they encounter a limited number of pediatric high-acuity situations [21,22]. Providers need to practice decisionmaking regularly to keep their skills fresh and to maintain confidence.
Currently, many providers at rural or community hospitals are underprepared to deal with pediatric trauma. Even trainees at academic trauma centers may have limited exposure or focused training to become proficient at managing pediatric trauma cases. Emergency care physicians are required to take the ATLS course, but only a small percentage of this course is devoted to pediatrics [13], and consequently, there is a lack of opportunities for providers to practice these skills. Moreover, there is significant evidence that these skills decay over time, and it is expensive and time consuming for providers to travel to these in-person courses [23]. According to recent literature, current teaching methods in PEM are inadequate to properly train providers for infrequent, dangerous events, such as trauma [24]. Other studies have shown that skills related to Advanced Cardiac Life Support (ACLS), a certification for providers, decay within six months, while the course is only taken every two years [25]. Some training facilities attempt to refresh skills with high fidelity simulation, such as mock codes and mannequin training, which require extensive planning and coordination, as well as expensive technology that is generally inflexible to the presence of new training needs and the presence of expert facilitators, thus making it impractical to train providers who have limited time and resources on the latest and most important skills.
As abdominal trauma is the leading cause of unrecognized fatal injury in children, training focused on its management is essential. Most practitioners base their pediatric trauma management on knowledge from prior experiences, medical literature, courses, and advice from experts. Due to limited exposure to pediatric trauma cases, the decision-making skills of a practitioner for a traumatically injured child may not be second nature. Our CTA results may be used to help learners to think like an expert. This CTA provides more explanation for decisionmaking compared to existing literature on trauma management or courses. While these resources outline the steps in management, they do not explain the associated decision-making with each step. This CTA includes more information, such as sensory cues.
Learners may also turn to experts to ask questions about management of a child with blunt abdominal injury; however, experts may have become unconsciously aware of their automated skills over years of experience [9]. For an experienced practitioner, it may be second nature; however, a novice must synthesize new information to plan next steps and may risk falling into novice traps/potential errors. Automaticity may occur with tasks that are practiced repetitively, such as assigning roles and conducting a systematic primary and secondary survey. Although relying on automaticity to perform skills is efficient, it may interfere with teaching if experts are unable to articulate their thought processes.
To counteract limited education in pediatric trauma, our findings based on national experts in trauma management may be used as a comprehensive supplementary tool to teach decisionmaking skills (Tables 16-19). It may be used to efficiently teach practitioners about blunt abdominal trauma management by highlighting the key decision-making points and novice traps/potential errors as elicited from experts. The tables outline the steps in the trauma process, associated assessment steps, decisions, things to avoid and suggestions to improve the process for pediatric blunt abdominal trauma management. Many of these decisions are shared by all mechanisms of a traumatic injury until primary and secondary surveys are completed.
Application of our CTA results can effectively decrease training time and increase proficiency for learners by highlighting important decisions outlined by experts. The CTA provides a pathway for practitioners to improve their decision-making skills when managing a patient with pediatric blunt abdominal trauma. These results may also be used as a script for simulation or used to develop an assessment tool or algorithm, for example, creating a decision-making algorithm to teach learners how to effectively make clinical decisions for management of pediatric blunt abdominal trauma. A future study may involve researching the efficacy of such a tool for teaching in a randomized trial and/or test the validity of such an assessment tool. Alternatively, our findings may be utilized as a blueprint for serious video gaming.

Limitations
Using CTA to create such a framework is novel to the field of pediatrics, emergency medicine, and specifically trauma management. Moreover, management of blunt abdominal trauma is a more dynamic process compared to prior studies that applied CTA to procedural and surgical skills. Although we had a small sample size of seven for our CTA, we believe data saturation was ensured through a purposive/theoretical sampling of experts in PEM and pediatric trauma surgery and by gaining a comprehensive understanding of the dimensions of the concepts and themes that emerged from the CTA. Watling and Lingard [26] argue that data saturation is easier to achieve with a theoretical sampling than a convenience sampling. Furthermore, our results were strengthened by external validation of our study through a nationwide survey of experts from PEM and pediatric trauma surgeons. We found a lack of national data on the use of FAST examinations for blunt abdominal trauma in children, highlighting the need for future studies in this area.

Conclusions
CTA may be applied to the management of pediatric blunt abdominal injury to outline the clinical decision-making process and more specifically to identify critical steps and potential errors to serve as a framework for education in PEM and pediatric trauma surgery. This tool can help train learners to think like an expert and it will also allow experts to share key teaching points with trainees. The findings from this study may be used as a script for simulation or serious video gaming. Future studies may apply CTA to other types of traumatic injuries and/or involve an interdisciplinary approach.

Additional Information Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: Funded by Children's Mercy Pediatric Emergency Medicine Department. The authors have no financial relationships relevant to this article to disclose. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.