Emergency Severity Index Research Articles

Triage physicians in an academic emergency department: Impact on resident education.

Overcrowding in emergency departments (EDs) in the United States has been linked to worse patient outcomes. Implementation of countermeasures such as a physician-in-triage (PIT) system have improved patient care and decreased wait times. The purpose of this study was to evaluate how a PIT system affects medical resident education in an academic ED. This was a retrospective observational comparison of resident metrics at a single-site, urban, academic ED before and after implementing a PIT system. Resident metrics of average emergency severity index (ESI), patients-per-hour, and in-training-examination scores were measured before and six months after the implementation of the PIT system. In total, 18,231 patients were evaluated by all residents in the study period before PIT implementation compared to 17,008 in the study period following PIT implementation. The average ESI among patients evaluated by residents decreased from 3.00 to 2.68 (p<0.01, 95% confidence interval [CI]= 0.31 to 0.33), while average resident patient-per-hour rate decreased from 1.41 to 1.32 (p<0.01, 95% CI= 0.05 to 0.13] and ITE scores saw no statistically significant change of 76.11 to 78.26 (p=0.26, 95% CI= -5.75 to 1.45). While these differences are statistically significant, they are likely not clinically significant. Our implementation of PIT system at one academic medical center minimally increased the acuity and minimally decreased the number of patients that residents see. This suggested that in our center, a PIT program did not detract from ED resident clinical education. However, further research with alternative markers in multiple centers is needed.

Стационарлық жедел жәрдем бөлімшелерінде қолданылатын триаж шкалалары (әдебиетке шолу)

The article describes different types of sorting scales used in inpatient emergency departments in different countries. Triage scales are designed to optimize patient waiting times according to the severity of their medical condition in order to treat the most intense symptom as quickly as possible and reduce the negative impact on the prognosis of a long delay before starting treatment. The aim of the study is to analyze the literature data describing the scales for triage in the emergency department, their comparative characteristics, and assessment of their reliability and reliability. Material and methods. An online literature research was conducted in databases such as Pubmed, Web of Science and Scopus, as well as on-line publications in Russian and English languages. The following terms were used to search for relevant literature sources: emergency department, triage, emergency severity index. A total of 813 literary sources were found, 37 of which were selected for further analysis. Inclusion criteria: studies performed in people, published in English, Russian, as well as full versions of articles. Exclusion criteria: summary of reports, newspaper publications, personal messages. Results and discussion. Analysis of the literature data allows us to judge a variety of approaches and systems of medical sorting of patients requiring emergency medical care, especially in cases of mass disasters and injuries. Analysis of existing sorting scales has shown that these scales are fivestep and adapted to the region and the health system, but there is no single universal scale. Conclusion. The introduction of structured triage by specially trained medical personnel in the emergency Department helps to accurately identify patients whose lives are at risk, especially during periods of insufficient treatment effectiveness. Therefore, five-level triage systems are recommended by national and international societies of emergency medicine. Keywords: triage, system triage, emergency severity index, emergency department.

Riot-related injuries managed at a hospital in Beirut, Lebanon

Between October 2019 and February 2020, massive crowds protested in Lebanon against economic collapse. Various less than lethal weapons including riot control agents and rubber bullets were used by law enforcement, which led to several traumatic and chemical injuries among victims. This study describes the clinical presentation, management, outcome, and healthcare costs of injuries. A retrospective review of the hospital records of all the casualties presenting to the Emergency Department of the American University of Beirut Medical Center between October 17th, 2019, and February 29th, 2020, was conducted. A total of 313 casualties were evaluated in the ED, with a mean age of 30.2 +/- 9.6years and a predominance of males (91.1%). Most were protestors (71.9%) and arrived through EMS (43.5%) at an influx rate of one patient presenting every 2.7-8min. Most patients (91.1%) presented with an Emergency Severity Index of 3. Most patients (77.6%) required imaging with 10% having major findings including fractures and hemorrhages. Stones, rocks, and tear gas canisters (30.7%) were the most common mechanism of injury. Musculoskeletal injuries were most common (62.6%), followed by lacerations (44.7%). The majority (93.3%) were treated and discharged home and 3.2% required hospital admission, with 2.6% requiring surgery. Less-than-lethal weapons can cause severe injuries and permanent morbidity. The use of riot control agents needs to be better controlled, and users need to be well trained in order to avoid misuse and to lessen the morbidity, mortality, and financial burden.

Evaluation of Patients with Suspicion of COVID-19 in Pediatric Emergency Department.

Objectives:Coronavirus disease 2019 (COVID-19) have different clinical presentations in children. Most symptomatic children with suspicion of COVID-19 have fever and respiratory symptoms. In this retrospective study, we aimed to describe demographic features, clinical characteristics, and outcomes of confirmed and probable COVID-19 patients admitted to our pediatric emergency department (ED).Methods:We identified 135 children (aged 1 month-18 years) with suspicion of the COVID-19 who were admitted to our ED between March 11 and May 12, 2020. The urgency of patients was evaluated according to their Pediatric Assessment Triangle (PAT) and Emergency Severity Index (ESI) scores. Patients were divided into two groups as confirmed cases (Group 1) and probable cases (Group 2). Clinical, laboratory, radiologic features, and the disease severity of patients were analyzed.Results:According to PAT evaluation, 82 patients (65.6%) were non-urgent. The most frequent ESI triage category level was 3 (n=102, 76.1%). Forty-one (30.4%) patients were identified as laboratory-confirmed cases. Fifty-five (40.7%) patients were between 28 days and 4 years of age. Fever with cough was the most frequent symptoms at the onset of illness in COVID-19 positive patients (n=16, 39%). Sixty-four (47.4%) patients had mild disease and 40 (29.6%) patients had comorbidities. In Group 1, neutropenia was significantly higher than Group 2 (p=0.024). Mean procalcitonin and erythrocyte sedimentation rate levels of Group 2 were significantly higher than Group 1 (p=0.012 and p=0.028, respectively). Twenty-eight of 51 patients had chest computed tomography findings which were compatible with COVID-19. Fifty-one (37.8%) patients were discharged from ED, 81 (60%) were admitted to the ward, and 3 (2.2%) were admitted to the pediatric intensive care unit.Conclusion:During our study, we confirmed the diagnosis of 45 of 135 probable cases with the SARS-CoV-2 polymerase chain reaction test. Among confirmed COVID-19 cases, most of our patients had mild or moderate disease. The clinic of only confirmed three patients was classified as severe disease, and we had no critically ill patient.

Establishing Research Priorities for the Emergency Severity Index Using a Modified Delphi Approach

IntroductionThe accuracy of an initial ED triage decision has been reported to drive the clinical trajectory for ED patients, and, therefore, this assessment is critical to patient safety. The Emergency Severity Index—a 5-point score assigned by a triage nurse and based on disease acuity, patient potential for decompensation, and anticipated resource use—is used both in the United States and internationally. In the US, the Emergency Severity Index is used by up to 94% of the academic medical center emergency departments. In 2020, the Emergency Nurses Association acquired the intellectual property rights to the Emergency Severity Index and is responsible for its maintenance and improvement. ObjectiveThe purpose of this study was to establish a research agenda for the improvement of individual and institutional understanding and use of the Emergency Severity Index. MethodsModified Delphi process was used with 3 rounds of data collection. ResultsRound 1 yielded 112 issues, which were collapsed into 18 potential research questions in 4 general categories: education and training (6 questions), workplace environment (3 questions), emergency care services (7 questions), and special populations (2 questions). These questions were used in round 2 to establish importance. Round 3 yielded a rank ordering of both categories and research questions. DiscussionThe research priorities as set through the use of this modified Delphi process align well with current gaps in the literature. Research in these areas should be encouraged to improve the understanding of educational, environmental, and process challenges to emergency nurses’ triage decisions and accuracy of Emergency Severity Index assignments.

Improving ED Emergency Severity Index Acuity Assignment Using Machine Learning and Clinical Natural Language Processing

IntroductionTriage is critical to mitigating the effect of increased volume by determining patient acuity, need for resources, and establishing acuity-based patient prioritization. The purpose of this retrospective study was to determine whether historical EHR data can be used with clinical natural language processing and machine learning algorithms (KATE) to produce accurate ESI predictive models. MethodsThe KATE triage model was developed using 166,175 patient encounters from two participating hospitals. The model was tested against a random sample of encounters that were correctly assigned an acuity by study clinicians using the Emergency Severity Index (ESI) standard as a guide. ResultsAt the study sites, KATE predicted accurate ESI acuity assignments 75.7% of the time compared with nurses (59.8%) and the average of individual study clinicians (75.3%). KATE’s accuracy was 26.9% higher than the average nurse accuracy (P <.001). On the boundary between ESI 2 and ESI 3 acuity assignments, which relates to the risk of decompensation, KATE’s accuracy was 93.2% higher, with 80% accuracy compared with triage nurses 41.4% accuracy (P <.001). DiscussionKATE provides a triage acuity assignment more accurate than the triage nurses in this study sample. KATE operates independently of contextual factors, unaffected by the external pressures that can cause under triage and may mitigate biases that can negatively affect triage accuracy. Future research should focus on the impact of KATE providing feedback to triage nurses in real time, on mortality and morbidity, ED throughput, resource optimization, and nursing outcomes.

Diagnosing Dyspneic Older Adult Emergency Department Patients: A Pilot Study.

Dyspnea is the second leading cause of US emergency department (ED) visits and an independent predictor of morbidity and mortality1 in older adult patients aged ≥65 years. Unfortunately, the diagnosis of the cause of dyspnea presents diagnostic challenges to emergency physicians2-4 that disproportionately affects older adults.5 One in 5 dyspneic older adults experience missed diagnosis in the ED2 and 21% are treated for ≥1 pneumonia, acute exacerbation of chronic obstructive pulmonary disease [COPD], and acute exacerbation of heart failure [HF].5 Importantly, some may have multiple causes of their dyspnea but accurate diagnosis remains critical.

Does crowding influence emergency department treatment time and disposition?

ObjectiveThe objective of this study was to determine whether crowding influences treatment times and disposition decisions for emergency department (ED) patients.MethodsWe conducted a retrospective cohort study at 2 hospitals from January 1, 2014, to July 1, 2014. Adult ED visits with dispositions of discharge, admission, or transfer were included. Treatment times were modeled by linear regression with log‐transformation; disposition decisions (admission or transfer vs discharge) were modeled by logistic regression. Both models adjusted for chief complaint, Emergency Severity Index (ESI), and 4 crowding metrics in quartiles: waiting count, treatment count, boarding count, and National Emergency Department Overcrowding Scale.ResultsWe included 21,382 visits at site A (12.9% excluded) and 29,193 at site B (15.0% excluded). Respective quartiles of treatment count increased treatment times by 7.1%, 10.5%, and 13.3% at site A (P < 0.001) and by 4.0%, 6.5%, and 10.2% at site B (P < 0.001). The fourth quartile of treatment count increased estimates of treatment time for patients with chest pain and ESI level 2 from 2.5 to 2.9 hours at site A (20 minutes) and from 3.0 to 3.3 hours at site B (18 minutes). Treatment times decreased with quartiles of waiting count by 5.6%, 7.2%, and 7.3% at site B (P < 0.001). Odds of admission or transfer increased with quartiles of waiting count by 8.7%, 9.6%, and 20.3% at site A (P = 0.011) and for the third (11.7%) and fourth quartiles (27.3%) at site B (P < 0.001).ConclusionsLocal crowding influenced ED treatment times and disposition decisions at 2 hospitals after adjusting for chief complaint and ESI.

Triage Accuracy and Its Association with Patient Factors Using Emergency Severity Index: Findings from United Arab Emirates.

IntroductionThe Ministry of Health and Prevention of the UAE acquired an electronic medical record system (Wareed) through which they incorporated the Emergency Severity Index as the standard triaging tool. This raised the need to review population dynamics and the accuracy of triage performed by the health-care providers utilizing the tool.ObjectiveThis research aimed to study demographics and dynamics of the population presenting to emergency departments (EDs) during 2018, evaluate the accuracy of triage assessment using comparative analysis techniques, and determine relationships between patient factors (severity of illness, age-group) and the accuracy of triage.MethodsThis was an observational study that aimed to ascertain findings from ED data over 1 year (January 2018–December 2018) and explore factors associated with reduced accuracy in acuity assignment. We employed comparative analysis to measure the level of agreement between standard guidelines and local findings.ResultsA total of 576,154 patients visited EDs in 2018, of which 54.4% were male. A statistically significant increase in length of stay with increasing severity of illness was observed (Kruskal–Wallis test). Overall triage accuracy was 41.6%, with a positive association with increasing severity of illness. We found a positive association between severity of illness and accuracy of triage (OR 0.14, p=0). We also found on logistic regression that the age-group 11–20 years had the highest probability of accurate triage acuity (R2=0.41, p=0).ConclusionConducted on a very large data set from the UAE, our study reflects upon population dynamics and triage accuracy distribution among different variables. This study paves the way for further in-depth analysis of factors that may impact triage accuracy within EDs, and utilizing a similar approach it can be replicated in other settings as well.

Emergency Department Mean Physician Time per Patient and Workload Predictors ED-MPTPP.

One key element for emergency department (ED) staff calculation is the mean physician time per patient (MPTPP) and its influencing factors. The aims of this study were measuring the MPTPP, identifying factors with significant influence on the MPTPP, and developing a model to predict the MPTPP. This study was a prospective trial conducted at the ED of a university hospital in Germany. The MPTPP was measured with a specifically developed app. The influence of different factors on MPTPP were first tested in univariate analysis. Then, all significant factors were used in a multivariant regression model to minimize collinearities and to develop a prediction model. In total, 202 patients treated by 32 different physicians were observed within one year. The MPTPP was 47 min (standard deviation: 34 min). Relevant factors influencing the MPTPP were treatment area, Emergency Severity Index (ESI) triage level, guiding symptom category, and physician level (all p < 0.001). This model predicted 45% of the variance in the MPTPP (p < 0.001), which corresponds to a large effect size. We developed an effective prediction model for ED MPTPP, resulting in an MPTPP of 47 min. Future studies are needed to validate our model, which could serve as a benchmark for other EDs where the MPTPP is not available.

Emergency department patients with weakness or fatigue: Can physicians predict their outcomes at the front door? A prospective observational study.

BackgroundGeneralized weakness and fatigue are underexplored symptoms in emergency medicine. Triage tools often underestimate patients presenting to the emergency department (ED) with these nonspecific symptoms (Nemec et al., 2010). At the same time, physicians’ disease severity rating (DSR) on a scale from 0 (not sick at all) to 10 (extremely sick) predicts key outcomes in ED patients (Beglinger et al., 2015; Rohacek et al., 2015). Our goals were (1) to characterize ED patients with weakness and/or fatigue (W|F); to explore (2) to what extent physicians’ DSR at triage can predict five key outcomes in ED patients with W|F; (3) how well DSR performs relative to two commonly used benchmark methods, the Emergency Severity Index (ESI) and the Charlson Comorbidity Index (CCI); (4) to what extent DSR provides predictive information beyond ESI, CCI, or their linear combination, i.e., whether ESI and CCI should be used alone or in combination with DSR; and (5) to what extent ESI, CCI, or their linear combination provide predictive information beyond DSR alone, i.e., whether DSR should be used alone or in combination with ESI and / or CCI.MethodsProspective observational study between 2013–2015 (analysis in 2018–2020, study team blinded to hypothesis) conducted at a single center. We study an all-comer cohort of 3,960 patients (48% female patients, median age = 51 years, 94% completed 1-year follow-up). We looked at two primary outcomes (acute morbidity (Bingisser et al., 2017; Weigel et al., 2017) and all-cause 1- year mortality) and three secondary outcomes (in-hospital mortality, hospitalization and transfer to ICU). We assessed the predictive power (i.e., resolution, measured as the Area under the ROC Curve, AUC) of the scores and, using logistic regression, their linear combinations.FindingsCompared to patients without W|F (n = 3,227), patients with W|F (n = 733) showed higher prevalences for all five outcomes, reported more symptoms across both genders, and received higher DSRs (median = 4; interquartile range (IQR) = 3–6 vs. median = 3; IQR = 2–5). DSR predicted all five outcomes well above chance (i.e., AUCs > ~0.70), similarly well for both patients with and without W|F, and as good as or better than ESI and CCI in patients with and without W|F (except for 1-year mortality where CCI performs better). For acute morbidity, hospitalization, and transfer to ICU there is clear evidence that adding DSR to ESI and/or CCI improves predictions for both patient groups; for 1-year mortality and in-hospital mortality this holds for most, but not all comparisons. Adding ESI and/or CCI to DSR generally did not improve performance or even decreased it.ConclusionsThe use of physicians’ disease severity rating has never been investigated in patients with generalized weakness and fatigue. We show that physicians’ prediction of acute morbidity, mortality, hospitalization, and transfer to ICU through their DSR is also accurate in these patients. Across all patients, DSR is less predictive of acute morbidity for female than male patients, however. Future research should investigate how emergency physicians judge their patients’ clinical state at triage and how this can be improved and used in simple decision aids.

Clinical Predictors of Emergency Department Revisits within 48 Hours of Discharge; a Case Control Study

Introduction:Emergency department (ED) revisits increase overcrowding and predicting which patients may need to revisit could increase patient safety. This study aimed to identify clinical variables that could be used to predict the probability of revisiting ED within 48 hours of discharge. Methods:A retrospective case-control study was conducted between July 2018 and January 2019 at the Emergency Medicine Department in Ramathibodi Hospital, Bangkok, Thailand. Patients who revisited the ED within 48 hours of discharge (case group) and patients who did not (control group) participated. The predictive factors for ED revisit were identified through multivariate logistic regression analysis. Results:The case group consisted of 372 patients, who revisited the ED within 48 hours, and the control group consisted of 1488 patients. The most common reason for revisiting the ED was recurring gastrointestinal illness, in 107 patients (28.76%). According to the multivariate data analysis , five factors influenced the probability of revisiting the ED: age of more than 60 years (p < 0.001, OR = 2.04, 95%CI: 1.51-2.77), initial Emergency Severity Index (ESI) triage level of 2 (p = 0.007, OR = 1.20, 95%CI: 0.93-1.56), ED stay duration of 4 hours or longer (p = 0.013, OR = 1.12, 95%CI: 0.87-1.44), body temperature of ≥37.5ºC on discharge (p = 0.034, OR = 1.34, 95%CI: 1.00-1.80), and pulse rate of less than 60 (OR = 1.55, 95%CI: 0.87-2.77) or more than 100 beats/minute (OR = 1.53, 95%CI: 1.10-2.11) (p = 0.011).Conclusion:According to the findings, the most important and independent predictive factor of ED revisit within 48 hours of discharge were, age ≥ 60 years, ESI triage level 2, ED length of stay ≥ 4 hours, temperature ≥ 37.5 C, and 60 > pulse rate ≥ 100 beats/minute.

Assessment of ED triage of anaphylaxis patients based on the Emergency Severity Index

To describe the emergency department (ED) triage of anaphylaxis patients based on the Emergency Severity Index (ESI), assess the association between ESI triage level and ED epinephrine administration, and determine characteristics associated with lower acuity triage ESI assignment (levels 3 and 4). We conducted a cohort study of adult and pediatric anaphylaxis patients between September 2010 and September 2018 at an academic ED. Patient characteristics and management were compared between Emergency Severity Index (ESI) triage level 1 or 2 versus levels 3 or 4 using logistic regression analysis. We adhered to STROBE reporting guidelines. A total of 1090 patient visits were included. There were 26 (2%), 515 (47%), 489 (45%), and 60 (6%) visits that were assigned an ESI triage level of 1, 2, 3, and 4, respectively. Epinephrine was administered in the ED to 53% of patients triaged ESI level 1 or 2 and to 40% of patients triaged ESI level 3 or 4. Patients who were assigned a lower acuity ESI level of 3 or 4 had a longer median time from ED arrival to epinephrine administration compared to those with a higher acuity ESI level of 1 or 2 (28min compared to 13min, p<.001). A lower acuity ESI level was more likely to be assigned to visits with a chief concern of hives, rash, or pruritus (OR 2.33 [95% CI, 1.20-4.53]) and less likely to be assigned to visits among adults (OR, 0.43 [0.31-0.60]), patients who received epinephrine from emergency medical services (OR 0.56 [0.38-0.82]), presented with posterior pharyngeal or uvular angioedema (OR, 0.56 [0.38-0.82]), hypoxemia (OR, 0.34 [0.18-0.64]), or increased heart (OR 0.83 [0.73-0.95]) or respiratory (OR 0.70 [0.60-0.82]) rates. Patients triaged to lower acuity ESI levels experienced delays in ED epinephrine administration. Adult and pediatric patients with skin-related chief concerns were more likely to be to be assigned lower acuity ESI levels. Further studies are needed to identify interventions that will improve ED anaphylaxis triage.

Concordance of triage performed with emergency severity index in the emergency departments of Rasoul Akram and Haft Tir hospitals

Introduction: Triage is prioritizing patients by disease severity in the shortest possible time. Proper triage increases the quality of patient care services, increases patient’s satisfaction, decreases patient waiting time, decreases mortality, and enhances emergency department efficiency. Objectives: The aim of this study was to evaluate the concordance of triage performed by nurses with ESI standard, at Rasoul Akram and Haft Tir hospitals, which are the most important educational centers of Iran University of Medical Sciences. Patients and Methods: This is a cross-sectional design study done in 2019. The study population is patients referred to Rasoul Akram and Haft Tir hospitals. A sample of 800 patients was selected. In this study, we determined how triage level were determined and compared with triaging by emergency medicine assistant. Data were analyzed by SPSS 23. Results: Out of 800 samples, the highest triage level was related to level three (79.1%) and levels two, four and one were in the next levels. 13.5% of the cases were not properly triaged according to the ESI standard and the adaptation coefficient (kappa) between the triage level specified in the patient file with the emergency severity index (ESI) triage level between physician and nurse was 58%, which was a significant difference (P&lt;0.001). The relationship between correct triage leveling and patients’ complaints was significant (P&lt;0.001). Conclusion: Nurses triage education about appropriate triage and ESI tool should be one of the most urgent topics and priorities of the studied hospitals. Over triage causes resources wasted and under triage may harm the patient’s health.

Descriptive Evaluation of Male Emergency Department Patients in the United States With Gonorrhea and Chlamydia.

IntroductionSexually transmitted infections are commonly tested for in the emergency department (ED), but diagnostic test results are often unavailable during the clinical encounter.MethodsWe retrospectively reviewed health records of 3,132 men ≥18 years that had an emergency department visit in northeast Ohio between April 18, 2014 and March 7, 2017. All subjects underwent testing for Neisseria gonorrhoeae and Chlamydia trachomatis. Independent t-tests and chi-square analyses were performed as well as multivariable regression analysis.ResultsOn univariable analysis, men with N gonorrhoeae and/or C trachomatis, compared with uninfected men, were younger (25.9 vs 32.4 years), more likely to be of Black race (91.7% vs 85.6%), less likely to be married (3.7% vs 10.2%), less likely to arrive to the ED by ambulance or police (1.7% vs 4.1%), and more likely to be diagnosed with a urinary tract infection (8.3% vs 3.7%), to be treated for gonorrhea and chlamydia in the ED (84.6% vs 54.9%), and to have higher emergency severity index (ESI) scores (3.8 vs 3.6) (P ≤ .03 for all). On urinalysis, men infected with N gonorrhoeae and/or C trachomatis had significantly more white blood cells (55.1 vs 20.9); more mucus (1.3 vs 1.2); higher leukocyte esterase (1.5 vs .4); fewer squamous epithelial cells (.6 vs 1.4); higher urobilinogen (1.1 vs .8); higher bilirubin (.09 vs .05); and more protein (.4 vs .3) (P ≤ .04).ConclusionsDemographic and urinalysis findings can be associated with an increased odds of men being infected with N gonorrhoeae and/or C trachomatis.

Care intensity of spontaneous intracranial hemorrhage: Effectiveness of the critical care resuscitation unit.

• The Critical Care Resuscitation Unit provides high care intensity. • The Critical Care Resuscitation Unit further reduces patients' blood pressure variability. • Older age, higher GCS, and mechanical ventilation prior to transfer are associated with need for higher care intensity. • Increased age and high acuity based on the Emergency Severity Index during ED stay are associated with high mortality. • High maximum systolic blood pressures, history of hypertension, and male gender are factors associated with AKI.

Assessment of Interrater Reliability of the Emergency Severity Index After Implementation in Emergency Departments in Jamaica Using a Learning Collaborative Approach

IntroductionIn 2016, the Ministry of Health in Jamaica selected the Emergency Severity Index as the triage tool to be used nationally. This study evaluated the effectiveness of this approach by assessing the interrater reliability among new users trained with minimal resources by 2 experienced trainers, 1 local and 1 international. MethodsA retrospective case series review was conducted within an online learning collaborative framework. After completion of the training, the participants from each of the 19 clinical sites were asked to submit 2 triage cases per month for blinded review by the expert trainers. The triage categories assigned by each reviewer were compared with those assigned by the newly trained Emergency Severity Index providers. A weighted kappa value was calculated to assess the degree of agreement between the sites and the expert trainers. ResultsA total of 166 cases were received over the study period. Participation in the learning collaborative was consistently below 50%. The interrater reliability between the expert trainers (κ = 0.48) as well as between each scorer and each accident and emergency department site (κSF = 0.33, κPT = 0.26) was low, although there was improvement over the study period. Incomplete triage documentation limited raters' ability to assign triage categories and assess interrater reliability. DiscussionDespite a rigorous implementation process, the interrater reliability of the Emergency Severity Index skills of Jamaican emergency nurses and doctors when compared with that of the 2 experts was poor. Several areas were identified for strengthening. Considerations for the implementation of the Emergency Severity Index in countries outside of the US were also discussed.

Prediction of emergency department resource requirements during triage: An application of current natural language processing techniques.

ObjectiveAccurate triage in the emergency department (ED) is critical for medical safety and operational efficiency. We aimed to predict the number of future required ED resources, as defined by the Emergency Severity Index (ESI) triage protocol, using natural language processing of nursing triage notes.MethodsWe constructed a retrospective cohort of all 265,572 consecutive ED encounters from 2015 to 2016 from 3 separate clinically heterogeneous academically affiliated EDs. We excluded encounters missing relevant information, leaving 226,317 encounters. We calculated the number of resources used by patients in the ED retrospectively and based outcome categories on criteria defined in the ESI algorithm: 0 (30,604 encounters), 1 (49,315 encounters), and 2 or more (146,398 encounters). A neural network model was trained on a training subset to predict the number of resources using triage notes and clinical variables at triage. Model performance was evaluated using the test subset and was compared with human ratings.ResultsOverall model accuracy and macro F1 score for number of resources were 66.5% and 0.601, respectively. The model had similar macro F1 (0.589 vs 0.592) and overall accuracy (65.9% vs 69.0%) compared to human raters. Model predictions had slightly higher F1 scores and accuracy for 0 resources and were less accurate for 2 or more resources.ConclusionsMachine learning of nursing triage notes, combined with clinical data available at ED presentation, can be used to predict the number of required future ED resources. These findings suggest that machine learning may be a valuable adjunct tool in the initial triage of ED patients.

Comparison of Hospital Triage Systems Using the Modified Trauma Severity Frailty Index vs Emergency Severity Index in Adult Trauma Patients

Danner, Omar K. MD, FACS, FCCM; Saberian, Sepehr K. BS, MS-2; Sola, Richard K. MD; Butler, Caroline K. MD; Nguyen, Jonathan K. DO; Udobi, Kahdi K. MBBS FACS; Childs, Ed W. MD, FACS Author Information

376 Crowding is the Strongest Predictor of Left Without Being Seen Risk in a Pediatric Emergency Department

Emergency Department (ED) patients who leave without being seen (LWBS) are associated with adverse safety, medico-legal, patient experience, and financial consequences. While patient-level risk of LWBS has been previously tied to demographic and acuity-related factors, there is limited research linking this to ED crowding scales in the pediatric setting. The objective of this study was to determine the association between LWBS risk and two crowding scales, the National Emergency Department Overcrowding Score (NEDOCS) and occupancy rate, in the pediatric ED setting. We performed a retrospective observational study on administrative and electronic health record (EHR) data for the ED of a single quaternary care children’s hospital. The hospital saw just over 50,000 ED visits in 2019 and is verified as a Level 1 Trauma Center. All patients who presented during the 14-month study period (March 2, 2018 through May 2, 2019) were included. NEDOCS and occupancy rate were calculated for each 15-minute interval and matched to patient arrival time. We performed multiple logistic regression analyses to demonstrate the relationship between patient-level LWBS risk and NEDOCS, as well as LWBS risk and occupancy rate, controlling for demographic and acuity-related characteristics drawn from the pre-arrival state. We performed a dominance analysis using McFadden’s pseudo-R2 in order to determine the relative importance of our crowding metrics in the logistic regression models. A total of 54,890 patient encounters were included in this analysis, 1.22% of whom LWBS. Odds ratio for LWBS risk per 10-point increase in NEDOCS at individual time of arrival was 1.30 (95% CI 1.27-1.33). Odds ratio for LWBS per 10% increase in occupancy rate at individual time of arrival was 1.23 (95% CI 1.21-1.25). Area under the curve was 86.8% (95% CI 85.6%-87.9%) for the NEDOCS model and 86.6% for the occupancy rate model (95% CI 85.4%-87.8%). There was no statistically significant difference between the AUC of the two models (p-value 0.19). Dominance analysis revealed that in each model, the most important variable out of 11 studied was its respective crowding metric; NEDOCS accounted for 55.6% of predicted variance in LWBS rate in the first model while occupancy rate accounted for 53.9% of the predicted variance in LWBS in the second model. Time of day, emergency severity index, and insurance type were the next most important variables for both models, representing 10.7%, 10.2%, and 9.7%, respectively, of predicted variance in LWBS in the NEDOCS model. In this single-center study, ED crowding was the single most important factor that determined a patient’s likelihood of LWBS in the pediatric ED, accounting for over half of predicted variance. This highlights the importance of mitigating operational stress in the ED in order to deliver safe and reliable care. Previous studies have attempted to determine which crowding metric is best suited to measure operational stress in the ED; here we offer additional evidence that the simple measure of occupancy rate is non-inferior to NEDOCS, at least in the pediatric setting. Because occupancy rate and NEDOCS are available in real time, both could serve as a monitor for individual LWBS risk.