The effect of emergency department crowding on clinically oriented outcomes

Learning in Temporary Teams: The Varying Effects of Partner Exposure by Team Member Role

In many workplaces, temporary teams convene to coordinate complex work, despite team members having not worked together before. Most related research has found consistent performance benefits when members of temporary teams work together multiple times (team familiarity). Recent work in this area broke new conceptual ground by instead exploring the learning and performance benefits that team members gain by being exposed to many new partners (partner exposure). In contrast to that new work that examined partner exposure between team members who are peers, in this paper, we extend this research by developing and testing theory about the performance effects of partner exposure for team members whose roles are differentiated by authority and skill. We use visit-level data from a hospital emergency department and leverage the ad hoc assignment of attendings, nurses, and residents to teams and the round-robin assignment of patients to these teams as our identification strategy. We find a negative performance effect of both nurses’ and resident trainees’ partner exposure to more attendings and of attendings’ and nurses’ exposure to more residents. In contrast, both attendings and residents experience a positive impact on performance from working with more nurses. The respective effects of residents working with more attendings and with more nurses is attenuated on patient cases with more structured workflows. Our results suggest that interactions with team members in decision-executing roles, as opposed to decision-initiating roles, is an important but often unrecognized part of disciplinary training and team learning.

Early Prediction of Intensive Care Admission in Emergency Department Patients With Asthma

BACKGROUND

Emergency physicians must choose whether patients with asthma are admitted to a hospital ward or a higher level of care, such as an intermediate care unit (IMC) or intensive care unit (ICU).

OBJECTIVE

This study aimed to determine which variables, available early during emergency department (ED) visits, are associated with IMC/ICU admission.

METHODS

In this retrospective chart review (records from 2015-2018), two trained abstractors, blinded to study hypothesis, abstracted data on predictor variables and disposition (ward vs. IMC/ICU). Predictor variables were defined explicitly and abstracted from the periods of ED arrival and after treatment with 7.5 mg nebulized albuterol. Distress was defined as tripod positioning or speaking in broken sentences. "Arrival" and "after treatment" scoring systems were derived based on adjusted odds ratios (aOR) for predictor variables. We performed analyses using SAS^Ⓡ, version 9.4 (SAS Institute).

RESULTS

Among 273 patients, 105 required admission to an IMC/ICU. At presentation, distress (aOR 2.1, 95% confidence interval [CI] 1.1-3.9), room air SpO₂ ≥95% (aOR 0.29, 95% CI 0.14-0.62), respiratory rate > 20 breaths/min (aOR 1.9, 95% CI 1.0-3.3), and retractions (aOR 1.9, 95% CI 1.1-3.3) were associated with IMC/ICU admission. After initial bronchodilator therapy, heart rate > 120 beats/min (aOR 7.1, 95% CI 2.0-25), room air SpO₂ ≥ 95% (aOR 0.15, 95% CI 0.07-0.34), and noninvasive ventilation (aOR 6.5, 95% CI 2.5-17) were associated with IMC/ICU admission. Both scoring systems stratified risk of IMC/ICU admission into low-risk (9-10%) and high-risk (70-100%) groups.

CONCLUSIONS

Combinations of predictor variables, available early in a patient's stay, stratify risk of admission to an IMC/ICU bed.

Improving Emergency Department Throughput Using Audit-and-Feedback With Peer Comparison Among Emergency Department Physicians

INTRODUCTION

We sought to determine if audit-and-feedback with peer comparison among emergency physicians is associated with improved emergency department (ED) throughput and decreased variation in physician performance.

METHODS

We implemented an audit-and-feedback with peer comparison tool at a single urban academic ED from March 1, 2013, to July 1, 2018. In the first study period, physicians received no reports. In the second period, they received daily reports. In the third period, they received daily, quarterly, and annual reports. Outcomes included patients per hour, admission rate, time to admission, and time to discharge.

RESULTS

A total of 272,032 patient visits and 36 ED physicians were included. The mean admission rate decreased 6.8%; the mean time to admission decreased 43.8 minutes; and the mean time to discharge decreased 40.6 minutes. Variation among physicians decreased for admission rate, time to admission, and time to discharge. Low-performing outliers showed disproportionately larger improvements in patients per hour, admission rate, time to admission, and time to discharge.

CONCLUSIONS

Automated peer comparison reports for academic emergency physicians was associated with lower admission rates, shorter times to admission, and shorter times to discharge at the departmental level, as well as decreased practice variation at the individual level.

A blueprint for building an emergency department quality improvement and patient safety committee

The field of quality improvement and patient safety (QIPS) has matured significantly in emergency medicine over the past decade. From standalone, strategically misaligned, and incoherently designed QIPS projects years ago, emergency department (ED) leaders have now recognized that developing a more robust QIPS infrastructure helps prioritize and organize projects for a greater likelihood of success and impact for patients and the system. This process includes the development of a well-defined, accountable, and supported departmental QIPS committee. This can be achieved effectively using a deliberate and structured approach, such as the one described by Harvard Business School Professor John Kotter in his seminal work, “Leading Change.” Herein, we present a blueprint using this framework and include practical examples from our experience developing a robust and successful ED QIPS committee and infrastructure. The steps include how to develop a “burning platform,” select a guiding coalition of leaders, develop a strategic vision and initiatives, recruit a volunteer army of members, enable actions for the committee, generate short-term successes, sustain the pace of change, and, finally, enable the infrastructure to support ongoing improvements. This road map can be replicated by ED teams of variable sizes and settings to structure, prioritize, and operationalize their QIPS activities and ultimately improve the outcomes of their patients.

Validity of Wait Time Complaints and Effect of Trainee Presence in an Ophthalmic Emergency Department

Purpose

Press Ganey (PG) surveys can provide a more comprehensive evaluation of specific patient complaints through input of comments. In this paper, we evaluate the validity of wait time complaints submitted via PG surveys in an ophthalmic emergency department (ED) and determine the effect of trainee presence on total time spent in the ED and negative PG comments.

Materials and Methods

This was a retrospective study evaluating PG reports for ED demographic, diagnostic, visit time and trainee involvement trends.

Results

Females made up 48.0% (4465) of overall ED patients yet submitted 62.8% (98) of the negative comments on PG surveys. More than a quarter (27.7%) of the negative comments pertained to wait time. Patient-reported times spent in the ED were significantly longer than actual total ED times (p < 0.001). Both reported and actual total ED times were significantly longer for patients who submitted wait time negative comments than those with other complaints (p < 0.001 vs p = 0.039, respectively). Patients seen by residents spent a significantly longer time in the ED than those seen by non-resident providers (p = 0.015), although self-reported total ED times did not differ significantly in these patients (p = 0.467).

Conclusion

Patients submitting complaints about wait time and those seen by treatment teams including residents spent a significantly longer time in the ED. Although PG survey respondents are not representative of the overall ED patient population, this highlights the need for academic institutions to seek methods to maximize efficiency in patient care without compromising trainee education.

Evaluation of a COVID-19 emergency department observation protocol

Objectives

Caring for patients with COVID-19 has resulted in a considerable strain on hospital capacity. One strategy to mitigate crowding is the use of ED-based observation units to care for patients who may have otherwise required hospitalization. We sought to create a COVID-19 Observation Protocol for our ED Observation Unit (EDOU) for patients with mild to moderate COVID-19 to allow emergency physicians (EP) to gather more data for or against admission and intervene in a timely manner to prevent clinical deterioration.

Methods

This was a retrospective cohort study which included all patients who were positive for SARS-CoV-2 at the time of EDOU placement for the primary purpose of monitoring COVID-19 disease. Our institution updated the ED Observation protocol partway into the study period. Descriptive statistics were used to characterize demographics. We assessed for differences in demographics, clinical characteristics, and outcomes between admitted and discharged patients. Multivariate logistic regression models were used to assess whether meeting criteria for the ED observation protocols predicted disposition.

Results

During the time period studied, 120 patients positive for SARS-CoV-2 were placed in the EDOU for the primary purpose of monitoring COVID-19 disease. The admission rate for patients in the EDOU during the study period was 35%. When limited to patients who met criteria for version 1 or version 2 of the protocol, this dropped to 21% and 25% respectively. Adherence to the observation protocol was 62% and 60% during the time of version 1 and version 2 implementation, respectively. Using a multivariate logistic regression, meeting criteria for either version 1 (OR = 3.17, 95% CI 1.34–7.53, p < 0.01) or version 2 (OR = 3.18, 95% CI 1.39–7.30, p < 0.01) of the protocol resulted in a higher likelihood of discharge. There was no difference in EDOU LOS between admitted and discharged patients.

Conclusion

An ED observation protocol can be successfully created and implemented for COVID-19 which allows the EP to determine which patients warrant hospitalization. Meeting protocol criteria results in an acceptable admission rate.

Association of County-Level Availability of Pediatricians With Emergency Department Visits

OBJECTIVES

The relationship between pediatrician availability and emergency department (ED) attendance is uncertain. We determined whether children in counties with more pediatricians had fewer ED visits.

METHODS

We conducted a cross-sectional study of all ED visits among children younger than 18 years from 6 states. We obtained ED visit incidences by county and assessed the relationship to pediatrician density (pediatricians per 1000 children). Possible confounders included state, presence of an urgent care facility in the county, urban-rural status, and quartile of county-level characteristics: English-speaking, Internet access, White race, socioeconomic status, and public insurance. We estimated county-level changes in incidence by pediatrician density adjusting for state and separately for all possible confounders.

RESULTS

Each additional pediatrician per 1000 children was associated with a 13.7% (95% confidence interval, -19.6% to -7.5%) decrease in ED visits in the state-adjusted model. In the full model, there was no association (-1.4%, 95% confidence interval, -7.2% to 4.8%). The presence of an urgent care, higher socioeconomic status score, urban status, and higher proportions of White race and nonpublic insurance were each associated with decreased ED visit rates.

CONCLUSIONS

Pediatrician density is not associated with decreased ED visits after adjusting for other county demographic factors. Increasing an area's availability of pediatricians may not affect ED attendance.

Can serum biomarkers be used to rule out significant intracranial pathology in emergency department patients with mild traumatic brain injury? A Systemic Review & Meta-Analysis

BACKGROUND

Interest has mounted into the use of objective clinical biomarkers for traumatic brain injury (TBI). This systematic review and meta-analysis aimed to synthesise the existing evidence investigating the use of serum & plasma biomarkers to exclude significant intracranial injuries seen on CT head scans in patients that present to ED with TBI.

METHODS

The primary outcome was to review the diagnostic accuracy (sensitivity & specificity) of S100B, GFAP and UCH-L1 to exclude significant intracranial pathology on CT head scan in adults presenting with TBI. Secondary outcomes investigated biomarker performance at different time points, in isolated TBI and multi-trauma and with pre-specified cut offs. Systematic searches were conducted on MEDLINE ® (via PubMed), Cochrane electronic databases and EMBASE from 1^st January 2000 until June 2020. Bias was assessed using QUADAS 2 tool. A narrative synthesis and meta-analysis were performed. PROSPERO registration number CRD42020212206.

RESULTS

After screening, 22 papers were included. The total number of patients with TBI was 9,416. There was significant variation regarding study design, population selection and the clinical threshold/decision rule for CT head request. The diagnostic accuracy of S100B as measured by the range of individual sensitivities and specificities were 63-100% and 5-58%, respectively. Individual sensitivities and specificities for GFAP were 67-100% and 0-89% and for UCH-L1 were 61-100% and 21-63.7% respectively. When measured within 3 hours individual sensitivities & specificities for S100B were 98-100% & 20-58% respectively. The quality of evidence for the primary outcome overall was low. The quality of evidence was low for all secondary outcomes apart from studies that used a pre-specified cut off for S100B which had a moderate strength of evidence.

CONCLUSION

The overall quality of evidence regarding the diagnostic accuracy of single biomarkers as a rule out for significant intracranial injury seen on CT head scans in ED patients with TBI is low. Based on current evidence, S100B is the only single biomarker with a validated clinical platform, pre-determined cut off threshold and moderate quality evidence; at this stage making it the biomarker of choice. More robust clinical outcome and economic impact data is required to support its incorporation into clinical decision tools.

Patients' experiences of attending emergency departments where primary care services are located: qualitative findings from patient and clinician interviews from a realist evaluation

BACKGROUND

Patient experience is an important outcome and indicator of healthcare quality, and patient reported experiences are key to improving quality of care. While patient experience in emergency departments (EDs) has been reported in research, there is limited evidence about patients' specific experiences with primary care services located in or alongside EDs. We aim to identify theories about patient experience and acceptability of being streamed to a primary care clinician in an ED.

METHODS

Using theories from a rapid realist review as a basis, we interviewed 24 patients and 106 staff members to generate updated theories about patient experience and acceptability of streaming to primary care services in EDs. Feedback from 56 stakeholders, including clinicians, policymakers and patient and public members, as well as observations at 13 EDs, also contributed to the development of these theories, which we present as a programme theory.

RESULTS

We found that patients had no expectations or preferences for which type of clinician they were seen by, and generally found being streamed to a primary care clinician in the ED acceptable. Clinicians and patients reported that patients generally found primary care streaming acceptable if they felt their complaint was dealt with suitably, in a timely manner, and when clinicians clearly communicated the need for investigations, and how these contributed to decision-making and treatment plans.

CONCLUSIONS

From our findings, we have developed a programme theory to demonstrate that service providers can expect that patients will be generally satisfied with their experience of being streamed to, and seen by, primary care clinicians working in these services. Service providers should consider the potential advantages and disadvantages of implementing primary care services at their ED. If primary care services are implemented, clear communication is needed between staff and patients, and patient feedback should be sought.

Implementation of a standardized pregnancy screening process to address gender disparities in radiology turn-around-time and ED length of stay

OBJECTIVE

The objective of this initiative was to quantify and intervene upon suspected gender disparities in CT turn-around-time and emergency department (ED) length of stay.

METHODS

This was a single-site before-after quality improvement initiative including patients aged 12-50 who underwent CT chest and/or abdomen/pelvis. The intervention included protocolization of the pregnancy screening process in triage. Primary outcomes included the difference between women of childbearing age and similarly aged men in regards to CT turn-around-time and ED length of stay. Pre- and post-intervention data were analyzed, including an "intensive intervention period" subanalysis.

RESULTS

CT turn-around-time for women of childbearing age was 19 min longer than for similarly aged men at baseline and did not change significantly post-intervention. ED length of stay was 27 min longer for women of childbearing age compared to similarly aged men at baseline and 7 min longer post-intervention, although this was still a significant difference. During the intensive intervention period, CT turn-around-time for women of childbearing age was 15 min longer than similarly aged men but the difference in ED length of stay of 10 min was no longer significant.

CONCLUSIONS

There is gender disparity in CT turn-around-time and ED length of stay in our ED, highlighting an important area for improvement to promote equitable care. A quality improvement initiative that aimed to protocolize pregnancy testing in triage did not show sustainable improvement in these outcomes but did result in increased pregnancy testing.

Prevalence of Prolonged Length of Stay in an Emergency Department in Urban Denmark: A Retrospective Health Records Repository Review

INTRODUCTION

Prolonged length of stay in emergency departments is associated with increased hospitalization, hospital-acquired pressure ulcers, medication errors, and mortality. In acute admissions in Denmark in 2018, 67% of patients experienced waiting time from arrival to examination. This study aimed to estimate the prevalence of prolonged length of stay (≥6 hours) and identify risk factors related to input, throughput, and output components.

METHODS

A retrospective health records repository review included 4743 patients admitted to a single urban emergency department in Denmark in January 2019. Data collected from the electronic health record system repository included demographic and organizational characteristics and were analyzed using descriptive statistics and logistic regression.

RESULTS

Among patients admitted in the study period, 31% had a prolonged length of stay of ≥6 hours. Prolonged length of emergency department stay was associated with being female (male odds ratio [OR], 0.86; 95% confidence interval [CI], 0.75-0.98), treatment by medical service (OR, 4.25, 95% CI, 3.63-4.98) vs surgical or injury, triage acuity of 2-Orange (OR, 1.45; 95% CI, 1.18-1.78) or 3-Yellow (OR, 1.47; 95% CI, 1.23-1.75) on a 5-level scale, evening (OR, 1.44; 95% CI, 1.24-1.66) or night (OR, 2.36; 95% CI, 1.91-2.91) arrival, ages 56 to 80 (OR, 1.79; 95% CI, 1.52-2.11) and >81 (OR, 2.40; 95% CI, 1.99-2.88) years, and hospital admission (OR, 1.19; 95% CI, 1.04-1.38) vs discharge from the emergency department to home.

DISCUSSION

Female, elderly, and medical patients were each identified as at-risk characteristics for ≥6-hour length of stay in the emergency department. Acute care patient pathways in the emergency department, particularly for evening and night, with guideline-based care and system level improvements in patient flow are warranted. Further research with larger populations is needed to identify and support interventions to decrease prolonged length of stay.

A comparative evaluation of the strengths of association between different emergency department crowding metrics and repeat visits within 72 hours

OBJECTIVE

We sought to compare strengths of association among multiple emergency department (ED) input, throughput and output metrics and the outcome of 72-h ED re-visits.

METHODS

This database analysis used healthcare administrative data from three urban, university-affiliated EDs in Calgary, Canada, calendar years 2010-2014. We used data from all patients presenting to participating EDs during the study period, and the primary analysis was performed on patients discharged from the ED. Regression models quantified the association between input, throughput and output metrics and the risk of return ED visit within 72 h of discharge from the index ED encounter. Strength of association between the crowding metrics and 72-h ED re-visits was compared using Akaike's Information Criterion.

RESULTS

The findings of this study are based on data from 845,588 patient encounters ending in discharge. The input metric with the strongest association with 72-h re-visits was median ED waiting time. The throughput metric with the strongest association with 72-h re-visits was the ED occupancy. The output metric with the strongest association with 72-h re-visits was the median inpatient boarding time.

CONCLUSION

Input, throughput and output metrics are all associated with 72-h re-visits. Delays in any of these operational phases have detrimental effects on patient outcomes. ED waiting time, ED occupancy, and boarding times are the most meaningful input, throughput and output metrics. These should be the preferred metrics for quantifying ED crowding in research and quality improvement efforts, and for clinicians to monitor ED crowding in real time.

Emergency department use and length of stay by younger and older adults: Nottingham cohort study in the emergency department (NOCED)

BACKGROUND

Younger and older adults attending the Emergency Department (ED) are a heterogeneous population. Longer length of ED stay is associated with adverse outcomes and may vary by age.

AIMS

To evaluate the associations between age and (1) clinical characteristics and (2) length of ED stay among adults attending ED.

METHODS

The NOttingham Cohort study in the Emergency Department (NOCED)-a retrospective cohort study-comprises new consecutive ED attendances by adults ≥ 18 years, at a secondary/tertiary care hospital, in 2019. Length of ED stay was dichotomised as < 4 and ≥ 4 h. The associations between age and length of ED stay were analysed by binary logistic regression and adjusted for socio-demographic and clinical factors including triage acuity.

RESULTS

146,636 attendances were analysed; 75,636 (51.6%) resulted in a length of ED stay ≥ 4 h. Attendances of adults aged 65 to 74 years, 75 to 84 years and ≥ 85 years, respectively, had an increased risk (odds ratio (95% confidence interval) of length of ED stay ≥ 4 h of 1.52 (1.45-1.58), 1.65 (1.58-1.72), and 1.84 (1.75-1.93), compared to those of adults 18 to 64 years (all p < 0.001). These findings remained consistent in the subsets of attendances leading to hospital admission and those leading to discharge from ED.

DISCUSSION AND CONCLUSION

In this real-world cohort study, older adults were more likely to have a length of ED stay ≥ 4 h, with the oldest old having the highest risk. ED target times should take into account age of attendees.

Accuracy of the Traditional COVID-19 Phone Triaging System and Phone Triage-Driven Deep Learning Model

Objectives:

During the COVID-19 pandemic, a quick and reliable phone-triage system is critical for early care and efficient distribution of hospital resources. The study aimed to assess the accuracy of the traditional phone-triage system and phone triage-driven deep learning model in the prediction of positive COVID-19 patients.

Setting:

This is a retrospective study conducted at the family medicine department, Cairo University.

Methods:

The study included a dataset of 943 suspected COVID-19 patients from the phone triage during the first wave of the pandemic. The accuracy of the phone triaging system was assessed. PCR-dependent and phone triage-driven deep learning model for automated classifications of natural human responses was conducted.

Results:

Based on the RT-PCR results, we found that myalgia, fever, and contact with a case with respiratory symptoms had the highest sensitivity among the symptoms/ risk factors that were asked during the phone calls (86.3%, 77.5%, and 75.1%, respectively). While immunodeficiency, smoking, and loss of smell or taste had the highest specificity (96.9%, 83.6%, and 74.0%, respectively). The positive predictive value (PPV) of phone triage was 48.4%. The classification accuracy achieved by the deep learning model was 66%, while the PPV was 70.5%.

Conclusion:

Phone triage and deep learning models are feasible and convenient tools for screening COVID-19 patients. Using the deep learning models for symptoms screening will help to provide the proper medical care as early as possible for those at a higher risk of developing severe illness paving the way for a more efficient allocation of the scanty health resources.

Prospective comparison of AMB, GAP AND START scores and triage nurse clinical judgement for predicting admission from an ED: a single-centre prospective study

BACKGROUND

It is postulated that early determination of the need for admission can improve flow through EDs. There are several scoring systems which have been developed for predicting patient admission at triage, although they have not been directly compared. In addition, it is not known if these scoring systems perform better than clinical judgement. Therefore, the aim of this study was to validate existing tools in predicting hospital admission during triage and then compare them with the clinical judgement of triage nurses.

METHODS

To conduct this prospective, single-centre observational study, we enrolled consecutive adult patients who presented between 30 September 2019 and 25 October 2019 at the ED of a large teaching hospital in Milan, Italy. For each patient, triage nurses recorded all of the variables needed to perform Ambulatory (AMB), Glasgow Admission Prediction (GAP) and Sydney Triage to Admission Risk Tool (START) scoring. The probability of admission was estimated by the triage nurses using clinical judgement and expressed as a percentage from 0 to 100 with intervals of 5. Nurse estimates were dichotomised for analysis, with ≥50% likelihood being a prediction of admission. Receiver operating characteristic curves were generated for accuracy of the predictions. Area under the curve (AUC) with 95% CI for each of the scores and for the nursing judgements was also calculated.

RESULTS

A total of 1710 patients (844 men; median age, 54 years (IQR: 34-75)) and 35 nurses (15 men; median age, 37 years (IQR: 33-48)) were included in this study. Among these patients, 310 (18%) were admitted to hospital from the ED. AUC values for AMB, GAP and START scores were 0.77 (95% CI: 0.74 to 0.79), 0.72 (95% CI: 0.69 to 0.75) and 0.61 (95% CI: 0.58 to 0.64), respectively. The AUC for nurse clinical judgement was 0.86 (95% CI: 0.84 to 0.89).

CONCLUSION

AMB, GAP and START scores provided moderate accuracy in predicting patient admission. However, all of the scores were significantly worse than the clinical judgement of the triage nurses.

Evaluation of feedback modalities and preferences regarding feedback on decision-making in a pediatric emergency department

OBJECTIVES

To compare pediatric emergency clinicians' attitudes toward three feedback modalities and assess clinicians' case-based feedback preferences.

METHODS

Electronic survey sent to pediatric emergency medicine (PEM) physicians and fellows; general pediatricians; and advanced practice providers (APPs) with nine questions exploring effectiveness and emotional impact of three feedback modalities: case-based feedback, bounce-back notifications, and biannual performance reports. Additional questions used a four-point ordinal agreement response scale and assessed clinicians' attitudes toward case review notification, case-based feedback preferences, and emotional support. Survey responses were compared by feedback modality using Pearson's chi-squared.

RESULTS

Of 165 eligible providers, 93 (56%) responded. Respondents agreed that case-based feedback was timely (81%), actionable (75%), prompted reflection on decision-making (92%), prompted research on current clinical practice (53%), and encouraged practice change (58%). Pediatric Emergency Care Applied Research Network (PECARN) performance reports scored the lowest on all metrics except positive feedback. No more than 40% of providers indicated that any feedback modality provided emotional support. Regarding case-based feedback, 88% of respondents desired email notification before case review and 88% desired feedback after case review. Clinicians prefer receiving feedback from someone with similar or more experience/training. Clinicians receiving feedback desire succinctness, supporting evidence, consistency, and sensitive delivery.

CONCLUSIONS

Case-based feedback scored highest of the three modalities and is perceived to be the most likely to improve decision-making and promote practice change. Most providers did not perceive emotional support from any feedback modality. Emotional safety warrants purposeful attention in feedback delivery. Critical components of case-based feedback include succinctness, supporting evidence, consistency, and sensitive delivery.

Real-time prediction of patient disposition and the impact of reporter confidence on mid-level triage accuracies: an observational study in Israel

AIM

The emergency department (ED) is the first port-of-call for most patients receiving hospital care and as such acts as a gatekeeper to the wards, directing patient flow through the hospital. ED overcrowding is a well-researched field and negatively affects patient outcome, staff well-being and hospital reputation. An accurate, real-time model capable of predicting ED overcrowding has obvious merit in a world becoming increasingly computational, although the complicated dynamics of the department have hindered international efforts to design such a model. Triage nurses' assessments have been shown to be accurate predictors of patient disposition and could, therefore, be useful input for overcrowding and patient flow models.

METHODS

In this study, we assess the prediction capabilities of triage nurses in a level 1 urban hospital in central Israeli. ED settings included both acute and ambulatory wings. Nurses were asked to predict admission or discharge for each patient over a 3-month period as well as exact admission destination. Prediction confidence was used as an optimisation variable.

RESULT

Triage nurses accurately predicted whether the patient would be admitted or discharged in 77% of patients in the acute wing, rising to 88% when their prediction certainty was high. Accuracies were higher still for patients in the ambulatory wing. In particular, negative predictive values for admission were highly accurate at 90%, irrespective of area or certainty levels.

CONCLUSION

Nurses prediction of disposition should be considered for input for real-time ED models.

Which health-related reasons lead to prehospital emergency care and how does subjective emergency status connect to subsequent care?

Objectives/Background

In many countries, the use of emergency medical services (EMS) increases steadily each year. At the same time, the percentage of life-threatening complaints decreases. To redesign the system, an assessment and consideration of the patients’ perspectives is helpful.

Methods

We conducted a paper-based survey of German EMS patients who had at least one case of prehospital emergency care in 2016. Four health insurance companies sent out the questionnaire to 1312 insured persons. We linked the self-reported data of 254 respondents to corresponding claims data provided by their health insurance companies. The analysis focuses a.) how strongly patients tend to call EMS for themselves and others given different health-related scenarios, b.) self-perceived health complaints in their own index case of prehospital emergency care and c.) subjective emergency status in combination with so-called “objective” characteristics of subsequent EMS and inpatient care. We report principal diagnoses of (1) respondents, (2) 57,240 EMS users who are not part of the survey and (3) all 20,063,689 inpatients in German hospitals. Diagnoses for group 1 and 2 only cover the inpatient stay that started on the day of the last EMS use in 2016.

Results

According to the survey, the threshold to call an ambulance is lower for someone else than for oneself. In 89% of all cases during their own EMS use, a third party called the ambulance. The most common, self-reported complaints were pain (38%), problems with heart and circulation (32%), and loss of consciousness (17%). The majority of respondents indicated that their EMS use was due to an emergency (89%). We could detect no or only weak associations between patients’ subjective urgency and different items for objective care.

Conclusion

Dispatchers can possibly optimize or reduce the disposition of EMS staff and vehicles if they spoke directly to the patients more often. Nonetheless, there is need for further research on how strongly the patients’ perceived urgency may affect the disposition, rapidness of the service and transport targets.

Extended Patient Alone Time in Emergency Department Leads to Increased Risk of 30-Day Hospitalization

OBJECTIVES

This study was conducted to describe patients at risk for prolonged time alone in the emergency department (ED) and to determine the relationship between clinical outcomes, specifically 30-day hospitalization, and patient alone time (PAT) in the ED.

METHODS

An observational cohort design was used to evaluate PAT and patient characteristics in the ED. The study was conducted in a tertiary academic ED that has both adult and pediatric ED facilities and of patients placed in an acute care room for treatment between May 1 and July 31, 2016, excluding behavioral health patients. Simple linear regression and t tests were used to evaluate the relationship between patient characteristics and PAT. Logistic regression was used to evaluate the relationship between 30-day hospitalization and PAT.

RESULTS

Pediatric patients had the shortest total PAT compared with all older age groups (86.4 minutes versus 131 minutes, P < 0.001). Relationships were seen between PAT and patient characteristics, including age, geographic region, and the severity and complexity of the health condition. Controlling for Charlson comorbidity index and other potentially confounding variables, a logistic regression model showed that patients are more likely to be hospitalized within 30 days after their ED visit, with an odds ratio (95% confidence interval) of 1.056 (1.017-1.097) for each additional hour of PAT.

CONCLUSIONS

Patient alone time is not equal among all patient groups. Study results indicate that PAT is significantly associated with 30-day hospitalization. This conclusion indicates that PAT may affect patient outcomes and warrants further investigation.

Emergency department-initiated buprenorphine protocols: A national evaluation

OBJECTIVE

Emergency department-initiated buprenorphine (BUP) for opioid use disorder is an evidence-based practice, but limited data exist on BUP initiation practices in real-world settings. We sought to characterize protocols for BUP initiation among a geographically diverse sample of emergency departments (EDs).

METHODS

In December 2020, we reviewed prestudy clinical BUP initiation protocols from all EDs participating in CTN0099 Emergency Department-INitiated bupreNOrphine VAlidaTION (ED-INNOVATION). We abstracted information on processes for identification of treatment-eligible patients, BUP administration, and discharge care.

RESULTS

All participating ED-INNOVATION sites across 22 states submitted protocols; 31 protocols were analyzed. Identification of treatment-eligible patients: Most EDs 22 (71%) relied on clinician judgment to determine appropriateness of BUP treatment with only 7 (23%) requiring decision support tools or diagnosis checklists. Before BUP initiation, 27 (87%) protocols required a documented Clinical Opiate Withdrawal Scale (COWS) score; 4 (13%) required a clinical diagnosis of withdrawal with optional COWS score. Twenty-seven (87%) recommended a minimum COWS score of 8 for ED-initiated BUP. BUP administration: Initial BUP dose ranged from 2-16 mg (mode = 4). For continued withdrawal symptoms, 27 (87%) protocols recommended an interval of 30-60 minutes between first and second BUP dose. Total BUP dose in the ED ranged from 8 to 32 mg. Discharge care: Twenty-eight (90%) protocols recommended a BUP prescription (mode 16 mg daily) at discharge. Naloxone prescription and/or provision was suggested in 23 (74%) protocols.

CONCLUSIONS

In this geographically diverse sample of EDs, protocols for ED-initiated BUP differed between sites. Future work should evaluate the association between this variation and patient outcomes.

Examination of the relationship between emergency department presentations and population mortality: a multicenter analysis of emergency department presentations during the COVID-19 pandemic

BACKGROUND

In the spring of 2020, Italy experienced a significant reduction in the number of emergency department (ED) presentations during the first wave of the COVID-19 pandemic. If ED access has an impact on patients' prognosis, such a reduction in ED presentations would be expected to correlate with a parallel increase in the mortality rate of the corresponding population. The aim of the present study was to evaluate the impact of reduced ED presentations on the all-cause mortality of the general population.

METHODS

Absolute and relative variation in ED accesses from March 1 to April 30 of both 2019 and 2020 in three hub hospitals in areas with different COVID-19 prevalence and age-standardized mortality data from January 1 to June 30 in 2019 and 2020 of the same areas were evaluated.

RESULTS

During March and April 2020, ED consults were decreased of approximately 50% in all three hospitals, as compared with the same months in 2019. There was a marked increase in cumulative mortality in Milan (high SARS-CoV2 infection spread zone) compared with the same period in 2019. In the other two municipalities (Ferrara and Perugia), which had intermediate and low levels of infection spread, the mortality in 2020 was not substantially changed from that of 2019.

CONCLUSIONS

Taking into account the increase in mortality due to SARS-CoV-2, reductions in ED access did not seem to affect death rates. If this finding will be confirmed, ED organization and access would need to be reconsidered.

Analysis of Plastic Surgery Consultations in a High-Volume Paediatric Emergency Department: A Quality Improvement Initiative

Introduction

Consult services influence emergency department (ED) workflow. Prolonged ED length of stay (LOS) correlates with ED overcrowding and as a consequence decreased quality of care and satisfaction of health team professionals. To improve management of paediatric ED patients requiring plastic and reconstructive surgery (PRS) expertise, current processes were analyzed.

Methods

Patient characteristics and metrics of PRS consultations in our paediatric ED were collected over a 3-month period. Data analysis was followed by feedback education intervention to ED and PRS staff. Data collection was then resumed and results were compared to the pre-intervention period.

Results

One hundred ninety-eight PRS consultations were reviewed, mean patient age was 6.3 years. Most common (52%) diagnoses were burns and hand trauma; 81% of PRS referrals were deemed appropriate; 25% of PRS consults were requested after hour with no differences in patient characteristics compared to regular hours; 60% of consultations involved interventions in the ED. Time between ED registration and PRS consultation request (116.5 minutes), quality of procedural sedation (52% rated inadequate), and overall ED LOS (289.2 minutes) were identified as main areas of concern and addressed during feedback education intervention. Emergency department LOS and quality of sedation did not improve in the post-intervention period.

Conclusion

The study provides detailed insights in the characteristics of PRS consultation in the paediatric ED population. Despite high referral appropriateness and education feedback intervention, significant inefficiencies were identified that call for further collaborative efforts to optimize quality of care for paediatric ED patients and improve satisfaction of involved healthcare professionals.

Using explainable machine learning to identify patients at risk of reattendance at discharge from emergency departments

Short-term reattendances to emergency departments are a key quality of care indicator. Identifying patients at increased risk of early reattendance could help reduce the number of missed critical illnesses and could reduce avoidable utilization of emergency departments by enabling targeted post-discharge intervention. In this manuscript, we present a retrospective, single-centre study where we created and evaluated an extreme gradient boosting decision tree model trained to identify patients at risk of reattendance within 72 h of discharge from an emergency department (University Hospitals Southampton Foundation Trust, UK). Our model was trained using 35,447 attendances by 28,945 patients and evaluated on a hold-out test set featuring 8847 attendances by 7237 patients. The set of attendances from a given patient appeared exclusively in either the training or the test set. Our model was trained using both visit level variables (e.g., vital signs, arrival mode, and chief complaint) and a set of variables available in a patients electronic patient record, such as age and any recorded medical conditions. On the hold-out test set, our highest performing model obtained an AUROC of 0.747 (95% CI 0.722–0.773) and an average precision of 0.233 (95% CI 0.194–0.277). These results demonstrate that machine-learning models can be used to classify patients, with moderate performance, into low and high-risk groups for reattendance. We explained our models predictions using SHAP values, a concept developed from coalitional game theory, capable of explaining predictions at an attendance level. We demonstrated how clustering techniques (the UMAP algorithm) can be used to investigate the different sub-groups of explanations present in our patient cohort.

Predicting inhospital admission at the emergency department: a systematic review

Background

ED crowding has potential detrimental consequences for both patient care and staff. Advancing disposition can reduce crowding. This may be achieved by using prediction models for admission. This systematic review aims to present an overview of prediction models for admission at the ED. Furthermore, we aimed to identify the best prediction tool based on its performance, validation, calibration and clinical usability.

Methods

We included observational studies published in Embase.com, Medline Ovid, Cochrane CENTRAL, Web of Science Core Collection or Google scholar, in which admission models were developed or validated in a general medical population in European EDs including the UK. We used the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) checklist to assess quality of model development. Model performance was presented as discrimination and calibration. The search was performed on 11 October 2020.

Results

In total, 18 539 articles were identified. We included 11 studies, describing 16 different models, comprising the development of 9 models and 12 external validations of 11 models. The risk of bias of the development studies was considered low to medium. Discrimination, as represented by the area under the curve ranged from 0.630 to 0.878. Calibration was assessed in seven models and was strong. The best performing models are the models of Lucke et al and Cameron et al. These models combine clinical applicability, by inclusion of readily available parameters, and appropriate discrimination, calibration and validation.

Conclusion

None of the models are yet implemented in EDs. Further research is needed to assess the applicability and implementation of the best performing models in the ED.

Systematic review registration number

PROSPERO CRD42017057975.

Mixed effect of increasing outflow of medical patients from an emergency department

Background and aim

Since 2014, the annual number of patients entering our emergency department (ED) has increased significantly. These were primarily Internal Medicine (IM) patients, and of these, 25–30% were admitted. The present governmental policy presents a deterrent to adding IM beds for these patients, and Emergency and IM departments cope with ever-increasing number of IM patients. We describe a quality improvement intervention to increase outflow of IM patients from the ED to the IM departments.

Methods

We conducted a quality improvement intervention at the Shaare Zedek Medical Center from 2014 to 2018. The first stage consisted of an effort to increase morning discharges from the IM departments. The second stage consisted of establishing a process to increase the number of admissions to the IM departments from the ED.

Results

Implementation of the first stage led to an increased morning discharge rate from a baseline of 2–4 to 18%. The second stage led to an immediate mean (± SD) morning transfer of 35 ± 7 patients to the medical departments (8–12 per department), providing significant relief for the ED. However, the additional workload for the IM departments’ medical and nursing staff led to a rapid decrease in morning discharges, returning to pre-intervention rates. Throughout the period of the new throughput intervention, morning admissions increased from 30 to > 70%, and were sustained. The number of patients in each department increased from 36 to 38 to a new steady state of 42–44, included constant hallway housing, and often midday peaks of 48–50 patients. Mean length of stay did not change. IM physician and nurse dissatisfaction led to increased number of patients being admitted during the evening and night hours and fewer during the morning.

Conclusion

We describe a quality improvement intervention to improve outflow of medical patients from the ED in the morning hours. The new ED practices had mixed effects. They led to less ED crowding in the morning hours but increased dissatisfaction among the IM department medical and nursing staff due to an increased number of admissions in a limited number of hours. The present governmental reimbursement policy needs to address hospital overcrowding as it relates to limited community healthcare beds and an aging population.

The impact of Medicaid expansion on emergency department wait times

OBJECTIVE

To estimate the impact of Medicaid expansion on emergency department (ED) wait times.

DATA SOURCES

We used 2012-2017 hospital-level secondary data from the CMS Hospital Compare data warehouse.

STUDY DESIGN

We used a state-level difference-in-differences approach to identify the impact of Medicaid expansion on four measures of ED wait times: time before being seen by a provider; time before being sent home after being seen by a provider; boarding time spent in the ED waiting to be discharged to an inpatient room; and the percentage of patients who left without being seen. We compared outcomes in states that expanded Medicaid with those in states that did not expand Medicaid.

DATA COLLECTION/EXTRACTION METHODS

Our sample included all US acute care hospitals with EDs in states that did not ever expand Medicaid or that fully expanded Medicaid in January of 2014.

PRINCIPAL FINDINGS

Medicaid expansion was associated with a 3.1-min increase (SE: 0.994, baseline mean: 30.8 min) in the time spent waiting to see an ED provider, a relative increase of 10%. Patients who were eventually sent home after being seen by a provider experienced a 7.5-min increase (SE: 1.8, baseline mean 142.1 min) in wait time. Boarding time rose by 3.8 min (SE 1.9, baseline mean 111.4 min). The percentage of patients who left without being seen rose by 0.3 percentage points (SE: 0.09, baseline mean 2.0), a relative increase of 15.3%.

CONCLUSIONS

This study provides multistate evidence that Medicaid expansion increased ED wait times for patients, indicating that ED crowding may have worsened post-expansion. Future work should aim to uncover the mechanisms through which insurance expansion increased ED wait times to provide policy direction.

Determinants of time-to-disposition in patients who underwent CT for pulmonary embolism: a retrospective study

Background

Pulmonary embolism (PE) is a common life-threatening medical emergency that needs prompt diagnosis and management. Providing urgent care is a key determinant of quality in the emergency department (ED) and time-based targets have been implemented to reduce length of stay and overcrowding. The study aimed to determine factors that are associated with having a time-to-disposition of less than 4 h in patients with suspected PE who underwent computed tomography pulmonary angiography (CT-PA) to confirm the diagnosis.

Methods

After obtaining approval from the ethics committee, we conducted a retrospective observational study by examining CT-PA scans that was performed to rule out PE in all adult patients presenting at the ED between January 2018 and December 2019. Demographic information and clinical information, as well as arrival and disposition times were collected from electronic health records. Multivariable regression analysis was used to identify the independent factors associated with meeting the 4-h target in the ED.

Results

In total, the study involved 232 patients (76 men and 156 women). The median length of stay in the ED was 5.2 h and the 4-h target was achieved in 37% of patients. Multivariable logistic regression analysis revealed that a positive CT-PA scan for PE was independently associated with meeting the four-hour target in the ED (odds ratio [OR]: 2.2; 95% CI: 1.1–4.8). Furthermore, Hemoptysis was the only clinical symptom that served as an independent factor associated with meeting the 4-h target in the ED (OR: 10.4; 95% CI: 1.2–90.8).

Conclusion

Despite the lower number of staff and higher volume of patients on weekends, patients who presented on weekends had shorter stays and were more likely to meet the 4-h target. Careful clinical assessment, prior to requesting a CT-PA scan, is crucial, since negative CT-PA scans may be associated with failure to meet the 4-h target.

Prediction of hospitalization using artificial intelligence for urgent patients in the emergency department

Timely assessment to accurately prioritize patients is crucial for emergency department (ED) management. Urgent (i.e., level-3, on a 5-level emergency severity index system) patients have become a challenge since under-triage and over-triage often occur. This study was aimed to develop a computational model by artificial intelligence (AI) methodologies to accurately predict urgent patient outcomes using data that are readily available in most ED triage systems. We retrospectively collected data from the ED of a tertiary teaching hospital between January 1, 2015 and December 31, 2019. Eleven variables were used for data analysis and prediction model building, including 1 response, 2 demographic, and 8 clinical variables. A model to predict hospital admission was developed using neural networks and machine learning methodologies. A total of 282,971 samples of urgent (level-3) visits were included in the analysis. Our model achieved a validation area under the curve (AUC) of 0.8004 (95% CI 0.7963-0.8045). The optimal cutoff value identified by Youden's index for determining hospital admission was 0.5517. Using this cutoff value, the sensitivity was 0.6721 (95% CI 0.6624-0.6818), and the specificity was 0.7814 (95% CI 0.7777-0.7851), with a positive predictive value of 0.3660 (95% CI 0.3586-0.3733) and a negative predictive value of 0.9270 (95% CI 0.9244-0.9295). Subgroup analysis revealed that this model performed better in the nontraumatic adult subgroup and achieved a validation AUC of 0.8166 (95% CI 0.8199-0.8212). Our AI model accurately assessed the need for hospitalization for urgent patients, which constituted nearly 70% of ED visits. This model demonstrates the potential for streamlining ED operations using a very limited number of variables that are readily available in most ED triage systems. Subgroup analysis is an important topic for future investigation.

Association of a Novel Index of Hospital Capacity Strain with Admission to Intensive Care Units

Rationale: Prior approaches to measuring healthcare capacity strain have been constrained by using individual care units, limited metrics of strain, or general, rather than disease-specific, populations.Objectives: We sought to develop a novel composite strain index and measure its association with intensive care unit (ICU) admission decisions and hospital outcomes.Methods: Using more than 9.2 million acute care encounters from 27 Kaiser Permanente Northern California and Penn Medicine hospitals from 2013 to 2018, we deployed multivariable ridge logistic regression to develop a composite strain index based on hourly measurements of 22 capacity-strain metrics across emergency departments, wards, step-down units, and ICUs. We measured the association of this strain index with ICU admission and clinical outcomes using multivariable logistic and quantile regression.Results: Among high-acuity patients with sepsis (n = 90,150) and acute respiratory failure (ARF; n = 45,339) not requiring mechanical ventilation or vasopressors, strain at the time of emergency department disposition decision was inversely associated with the probability of ICU admission (sepsis: adjusted probability ranging from 29.0% [95% confidence interval, 28.0-30.0%] at the lowest strain index decile to 9.3% [8.7-9.9%] at the highest strain index decile; ARF: adjusted probability ranging from 47.2% [45.6-48.9%] at the lowest strain index decile to 12.1% [11.0-13.2%] at the highest strain index decile; P < 0.001 at all deciles). Among subgroups of patients who almost always or never went to the ICU, strain was not associated with hospital length of stay, mortality, or discharge disposition (all P ≥ 0.13). Strain was also not meaningfully associated with patient characteristics.Conclusions: Hospital strain, measured by a novel composite strain index, is strongly associated with ICU admission among patients with sepsis and/or ARF. This strain index fulfills the assumptions of a strong within-hospital instrumental variable for quantifying the net benefit of admission to the ICU for patients with sepsis and/or ARF.

What is the impact of team triage as an intervention on waiting times in an adult emergency department? - A systematic review

AIM

To examine the impact of team triage on waiting times in adult emergency departments.

DESIGN

A systematic review using narrative analysis.

METHOD

Systematic review methodology, which included quantitative research papers consisting of randomized control trials, cohort or quasi-experimental studies. The PICO framework was used to formulate the question. Using a structured search, databases were used to source the research papers. Databases searched were Cochrane, CINAHL and MEDLINE. Twelve (12) research papers met the inclusion criteria. Each of the 12 papers were quality appraised using a recognised checklist. Data extraction was carried out and the findings were analysed using a narrative approach.

RESULTS

It was found that senior emergency doctors in triage alongside the triage nurse allows for more timely decision making and appropriate investigation orders. Early bed requesting or referral to specialist consultation were also found to improve waiting times. Reduced numbers of patients who leave without being seen and lower mortality rates were recorded when using team triage. Patient satisfaction is also improved by team triage.

CONCLUSION

Team triage improves waiting times in the emergency department.

Targeted HIV screening in the emergency department

Despite considerable improvement in human immunodeficiency virus (HIV) knowledge and treatment in the last 3 decades, the overall number of people living with HIV (PLHIV) is still rising with up to one quarter being unaware of their HIV status. Early HIV diagnosis and treatment prolongs life, reduces transmission, improves quality of life, and is a cost-effective public health intervention. The emergency department (ED) sees a large number of patients from marginalized and traditionally underserved populations in whom HIV is known to be more prevalent and who may not attend traditional services because of either cultural reasons or because of a chaotic lifestyle. This article discusses the two main approaches to screening; 'Opt-out' screening offers testing routinely in all clinical settings, and 'Targeted' screening offers testing to individuals presenting with indicator conditions. There are many studies of 'Opt-out' ED HIV screening in urban areas of high-HIV prevalence. However, little is known about the effectiveness of 'targeted' HIV screening, especially in areas of low prevalence. This review discusses the background to HIV screening in the ED and reviews the evidence around 'targeted' HIV screening in adult EDs in different HIV prevalence settings, concluding that targeted HIV screening at the ED can be impactful, cost-effective, and well accepted in the ED population, but its long-term implementation requires extra funding and increased staffing resource limiting its application in low resource setting. Despite most evidence being from areas of high-HIV prevalence, targeted screening might also be appropriate in low-HIV prevalence areas.

The influence of a fast-track service on case flow and client satisfaction in a high-volume veterinary emergency department

OBJECTIVE

To assess the benefit of a fast-track service in the emergency department of a large, high-volume veterinary hospital.

DESIGN

Prospective, observational, clinical study.

SETTING

Emergency department of an urban, tertiary referral veterinary hospital.

ANIMALS

All animals presented to the emergency department between April 1 and April 30 in 2017 and 2018 were eligible for inclusion. Only patients seen on days in 2017 corresponding to those days of 2018 during which the fast-track service was available were studied.

MEASUREMENT AND MAIN RESULTS

Triage case logs were collected and reviewed for April 2017 (prefast-track) and 2018 (fast-track). The fast-track service was launched as a pilot program in April 2018 to provide expedited care to low acuity patients presented to the emergency department. The median number of daily emergency department cases did not differ between 2017 (45, range 26-64) and 2018 (47, range 38-64; P = 0.3). The median time from presentation until first discussion with a doctor for low acuity cases was lower in April 2017 (29 min, range 1-163) than in April 2018 (24 min, range 1-100; P < 0.001). This reduction in wait time was observed despite a 40% increase in low acuity case presentations in 2018. Wait times for high acuity patients did not differ between study periods. The number of cases that left without being seen was higher in April 2017 compared to April 2018 (77 and 45 cases, respectively P < 0.001).

CONCLUSIONS

Implementation of a fast-track service reduced wait time for low acuity cases without adversely impacting wait times for sicker patients and led to a reduction in clients leaving without being seen. By introducing the fast-track service in a large volume veterinary hospital, limited resources can be distributed to improve speed of care, case flow, and client satisfaction in the emergency department.

Seeing Red: Blood Sample Hemolysis Is Associated with Prolonged Emergency Department Throughput

INTRODUCTION

Hemolyzed emergency department (ED) blood specimens impose substantial burdens on various aspects of delivering care. The ED has the highest incidence of hemolysis among hospital departments. This study assessed the association and potential impact of hemolyzed blood samples on patient throughput time using ED length of stay (LOS) as the primary outcome measure.

METHODS

This study was a secondary analysis of data collected during a performance improvement project aimed at reducing the incidence of hemolysis in ED blood specimens. The electronic medical record was queried for potassium orders and results and for key patient throughput time points. Throughput times were stratified according to hemolysis, ED disposition (admitted vs discharged), and Emergency Services Index (ESI) triage categorization. Two-tailed t tests were used to compare throughput times for patients with and without hemolysis.

RESULTS

Potassium values were reported for 11 228 patient visits. The mean ED LOS was 287 minutes for patients with nonhemolyzed samples and 349 minutes for patients who had hemolyzed samples, a mean delay of 62 minutes. The mean throughput time for discharged patients was 92 minutes shorter in the group without hemolysis (337 vs 429 minutes). The mean throughput time for admitted patients was 28 minutes shorter in the group without hemolysis (264 vs 292 minutes). The increased LOS for patients with a hemolyzed blood sample was independent of the most commonly encountered ESI levels.

CONCLUSION

Hemolysis of blood samples obtained in the ED is associated with prolonged patient throughput via delays in patient disposition, independent of various markers of acuity, such as the patients' ultimate disposition or triage categorization.

Automatic Classification of the Korean Triage Acuity Scale in Simulated Emergency Rooms Using Speech Recognition and Natural Language Processing: a Proof of Concept Study

BACKGROUND

Rapid triage reduces the patients' stay time at an emergency department (ED). The Korean Triage Acuity Scale (KTAS) is mandatorily applied at EDs in South Korea. For rapid triage, we studied machine learning-based triage systems composed of a speech recognition model and natural language processing-based classification.

METHODS

We simulated 762 triage cases that consisted of 18 classes with six types of the main symptom (chest pain, dyspnea, fever, stroke, abdominal pain, and headache) and three levels of KTAS. In addition, we recorded conversations between emergency patients and clinicians during the simulation. We used speech recognition models to transcribe the conversation. Bidirectional Encoder Representation from Transformers (BERT), support vector machine (SVM), random forest (RF), and k-nearest neighbors (KNN) were used for KTAS and symptom classification. Additionally, we evaluated the Shapley Additive exPlanations (SHAP) values of features to interpret the classifiers.

RESULTS

The character error rate of the speech recognition model was reduced to 25.21% through transfer learning. With auto-transcribed scripts, support vector machine (area under the receiver operating characteristic curve [AUROC], 0.86; 95% confidence interval [CI], 0.81-0.9), KNN (AUROC, 0.89; 95% CI, 0.85-0.93), RF (AUROC, 0.86; 95% CI, 0.82-0.9) and BERT (AUROC, 0.82; 95% CI, 0.75-0.87) achieved excellent classification performance. Based on SHAP, we found "stress", "pain score point", "fever", "breath", "head" and "chest" were the important vocabularies for determining KTAS and symptoms.

CONCLUSION

We demonstrated the potential of an automatic KTAS classification system using speech recognition models, machine learning and BERT-based classifiers.

High-efficiency Practices of Residents in an Academic Emergency Department: A Mixed-methods Study

BACKGROUND

Emergency department utilization and crowding is increasing, putting additional pressure on emergency medicine (EM) residency programs to train efficient residents who can meet these demands. Specific practices associated with resident efficiency have yet to be identified. The objective of this study was to identify practices associated with enhanced efficiency in EM residents.

METHODS

A mixed-methods study design was utilized to identify behaviors associated with resident efficiency. In Stage 1, eight EM faculty provided 61 efficiency behaviors during semistructured interviews, which were prioritized into eight behaviors by independent ranking. A total of 31 behaviors were tested, including additions from previous literature and the study team. In Stage 2, two 4-hour observations during separate shifts of 27 EM residents were performed to record minute-by-minute timing and frequency of each behavior. In Stage 3, the association between resident efficiency and each of the behaviors was estimated using multivariable regression models adjusted for training year and clustered on resident. The primary efficiency outcome was 6-month average relative value units/hour. A sensitivity analysis was performed using patients/hour.

RESULTS

Seven practices were positively associated with efficiency: average patient load, taking initial patient history with nurse present (number/hour, number/new patient), running the board (number/hour), conversations with other care team members (number/hour, % time), dictation use (number/hour, % time), smartphone text communication (number/hour, % time), and nonwork tasks (number/hour). Three practices were negatively associated with efficiency: visits to patient room (number/patient), conversations with attending physicians (% time), and reviewing electronic medical record (number/hour).

CONCLUSION

Several discrete behaviors were found to be associated with enhanced resident efficiency. These results can be utilized by EM residency programs to improve resident education and inform evaluations by providing specific, evidence-based practices for residents to develop and improve upon throughout training.

COVID-19: A Comprehensive Analysis of the Pandemic's Effect on an Emergency Department

OBJECTIVE

Coronavirus disease 2019 (COVID-19, caused by severe acute respiratory syndrome coronavirus 2 [SARS-CoV-2]) is a historic pandemic severely impacting health care. This study examines its early effect on a busy academic emergency department.

METHODS

A retrospective analysis of patients from an academic tertiary care Level I trauma, cardiac and stroke center's emergency department seeing an average of 54,000 adults and 21,000 pediatric patients per year. Total visits, reasons for patient visits, demographics, disposition, and length of stay were analyzed from January through July 2020 and compared with the same time period in the previous 2 y.

RESULTS

From March through July 2020 there were statistically significant decreases in the total number of patient visits (-47%) especially among pediatric (-73%) and elderly (-43%) patients and those with cardiovascular (-39%), neurological (-63%) complaints, headaches (-60%), back pain (-64%), abdominal pain (-51%), and minor trauma (-71%). There was, however, a significant increase in pulmonary complaints (+54%), as well as admissions (+32%), and length of stay (+40%).

CONCLUSIONS

There was a significant drop in overall patients and select groups early in the pandemic, while admissions and emergency department length of stay both increased. This has implications for future pandemic planning.

Performance evaluation of Emergency Department patient arrivals forecasting models by including meteorological and calendar information: A comparative study

The volume of daily patient arrivals at Emergency Departments (EDs) is unpredictable and is a significant reason of ED crowding in hospitals worldwide. Timely forecast of patients arriving at ED can help the hospital management in early planning and avoiding of overcrowding. Many different ED patient arrivals forecasting models have previously been proposed by using time series analysis methods. Even though the time series methods such as Linear and Logistic Regression, Autoregressive Integrated Moving Average (ARIMA), Seasonal ARIMA (SARIMA), Exponential Smoothing (ES), and Artificial Neural Network (ANN) have been explored extensively for the ED forecasting model development, the few significant limitations of these methods associated in the analysis of time series data make the models inadequate in many practical situations. Therefore, in this paper, Machine Learning (ML)-based Random Forest (RF) regressor, and Deep Neural Network (DNN)-based Long Short-Term Memory (LSTM) and Convolutional Neural network (CNN) methods, which have not been explored to the same extent as the other time series techniques, are implemented by incorporating meteorological and calendar parameters for the development of forecasting models. The performances of the developed three models in forecasting ED patient arrivals are evaluated. Among the three models, CNN outperformed for short-term (3 days in advance) patient arrivals prediction with Mean Absolute Percentage Error (MAPE) of 9.24% and LSTM performed better for moderate-term (7 days in advance) patient arrivals prediction with MAPE of 8.91% using weather forecast information. Whereas, LSTM model outperformed with MAPE of 8.04% compared to 9.53% by CNN and 10.10% by RF model for current day prediction of patient arrivals using 3 days past weather information. Thus, for short-term ED patient arrivals forecasting, DNN-based model performed better compared to RF regressor ML-based model.

Characteristics and outcomes of patients with an unscheduled return visit within 72 hours to the Paediatric Emergency Centre at a Private Tertiary Referral Hospital in Kenya

Introduction

Patients’ unscheduled return visits (URVs) to the paediatric emergency Centre (PEC) contribute to overcrowding and affect health service delivery and overall quality of care. This study assessed the characteristics and outcomes of paediatric patients with URVs (within 72 hours) to the PEC at a private tertiary hospital in Kenya.

Methods

We conducted a retrospective chart review of all URVs within 72 hours among paediatric patients aged ≤15 years between 1 July and 31 December 2018 at the tertiary hospital in Nairobi, Kenya.

Results

During the study period, 1.6% (n=172) of patients who visited the PEC returned within 72 hours, with 4.7% revisiting the PEC more than once. Patients’ median age was 36 months (interquartile range: 42 months); over half were male (51.7%), 55.8% were ambulatory and 84.3% were insured. In addition, 21% (n=36) had chronic diseases and 7% (n=12) had drug allergies. Respiratory (59.5%) and gastrointestinal (21.5%) tract infections were the most common diagnoses. Compared with the first visit, more patients with URVs were classified as urgent (1.7% vs. 5.2%) and were non-ambulatory (44.2% vs. 49.5%, p=<0.001); 18% of these patients were admitted. Of these 58% were male, 83.9% were aged 0–5 years, 12.9% were classified as urgent, 64.5% had respiratory tract infections and 16.1% had gastrointestinal tract infections. Being admitted was associated with patient acuity (p=0.004), laboratory tests (p=<0.001) and ambulatory status (p=0.041).

Conclusion

The URV rate is low in our setting. Patients who returned to the PEC within 72 hours tended to be male, under 5 years old and insured. Many were non-urgent cases with diagnoses of respiratory and gastrointestinal tract infections. The findings suggest that some URVs were necessary and may have contributed to better care and improved outcomes while others highlight a need for effective patient education and comprehensive initial assessment.

Morning Discharges and Patient Length of Stay in Inpatient General Internal Medicine

BACKGROUND

Many initiatives seek to increase the number of morning hospital discharges to improve patient flow, but little evidence supports this practice.

OBJECTIVE

To determine the association between the number of morning discharges and emergency department (ED) length of stay (LOS) and hospital LOS in general internal medicine (GIM).

DESIGN, SETTING, AND PARTICIPANTS

Multicenter retrospective cohort study involving all GIM patients discharged between April 1, 2010, and October 31, 2017, at seven hospitals in Ontario, Canada.

MAIN MEASURES

The primary outcomes were ED LOS and hospital LOS, and secondary outcomes were 30-day readmission and in-hospital mortality. The number of morning GIM discharges (defined as the number of patients discharged alive between 8:00 AM and 12:00 PM) on the day of each hospital admission was the primary exposure. Multivariable regression models were fit to control for patient characteristics and situational factors, including GIM census.

RESULTS

The sample included 189,781 patient admissions. In total, 36,043 (19.0%) discharges occurred between 8:00 AM and 12:00 PM. The average daily number of morning discharges and total discharges per hospital was 1.7 (SD, 1.4) and 8.4 (SD, 4.6), respectively. The median ED LOS was 14.5 hours (interquartile range [IQR], 10.0- 23.1), and the median hospital LOS was 4.6 days (IQR, 2.4-9.0). After multivariable adjustment, there was not a significant association between morning discharge and hospital LOS (adjusted rate ratio [aRR], 1.000; 95% CI, 0.996-1.000; P = .997), ED LOS (aRR, 0.999; 95% CI, 0.997-1.000; P = .307), 30-day readmission (aRR, 1.010; 95% CI, 0.991-1.020; P = .471), or in-hospital mortality (aRR, 0.967; 95% CI, 0.920-1.020; P = .183). The lack of association between morning discharge and LOS was generally consistent across all seven hospitals. At one hospital, morning discharge was associated with a 1.9% shorter ED LOS after multivariable adjustment (aRR, 0.981; 95% CI, 0.966-0.996; P = .013).

CONCLUSIONS

The number of morning discharges was not significantly associated with shorter ED LOS or hospital LOS in GIM. Our findings suggest that increasing the number of morning discharges alone is unlikely to substantially improve patient throughput in GIM, but further research is needed to determine the effectiveness of specific interventions.

Cost-minimization modeling of venous thromboembolism diagnostics: performing limited compression ultrasound in primary health care reduces costs compared to referring patients to a hospital

Background

The aim of this retrospective study was to determine whether diagnosing a deep venous thrombosis (DVT) in primary health care using limited compression ultrasound (LCUS) can save resources compared to referring these patients to hospital. According to the current literature, LCUS is as safe as a standard protocol based on a whole-leg ultrasound (US).

Methods

We created a standardized patient for this cost-analysis model based on 76 patients that were referred to hospital for a suspected DVT. Travel distance to the health care centre and hospital was calculated based on the home address. Hospital costs were acquired from the hospital price list and Finnish legislation. Time spent in the hospital was retrieved from hospital statistics. Time spent in the health care centre and travelling were estimated and monetized based on average salary. The cost of participating physicians attending a US training course was estimated based on the national average salary of a general practitioner as well as the course participation fee. A cost-minimization modeling was performed for this standardized patient comparing the total costs, including private and public costs, of standard and LCUS strategies.

Results

The total costs per patient of standard and LCUS pathways were 1151.52€ and 301.94€ [difference 849.59€ (95% CI 800.21€–898.97€, p < 0.001)], respectively. The real-life costs of these strategies, considering that some patients are probably referred to hospital every year and including training costs, are 1151.53€ and 508.69€ [difference 642.84€ (95% CI 541.85€–743.82€)], respectively.

Conclusion

Using LCUS in diagnosing DVT in primary health care instead of referring these patients to the hospital is shown to save a significant amount of public and private resources.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13089-021-00227-5.

Potential impacts of general practitioners working in or alongside emergency departments in England: initial qualitative findings from a national mixed-methods evaluation

OBJECTIVES

To explore the potential impacts of introducing General Practitioners into Emergency Departments (GPED) from the perspectives of service leaders, health professionals and patients. These 'expectations of impact' can be used to generate hypotheses that will inform future implementations and evaluations of GPED.

DESIGN

Qualitative study consisting of 228 semistructured interviews.

SETTING

10 acute National Health Service (NHS) hospitals and the wider healthcare system in England. Interviews were undertaken face to face or via telephone. Data were analysed thematically.

PARTICIPANTS

124 health professionals and 94 patients and carers. 10 service leaders representing a range of national organisations and government departments across England (eg, NHS England and Department of Health) were also interviewed.

RESULTS

A range of GPED models are being implemented across the NHS due to different interpretations of national policy and variation in local context. This has resulted in stakeholders and organisations interpreting the aims of GPED differently and anticipating a range of potential impacts. Participants expected GPED to affect the following areas: ED performance indicators; patient outcome and experience; service access; staffing and workforce experience; and resources. Across these 'domains of influence', arguments for positive, negative and no effect of GPED were proposed.

CONCLUSIONS

Evaluating whether GPED has been successful will be challenging. However, despite uncertainty surrounding the direction of effect, there was agreement across all stakeholder groups on the areas that GPED would influence. As a result, we propose eight domains of influence that will inform our subsequent mixed-methods evaluation of GPED.

TRIAL REGISTRATION NUMBER

ISRCTN51780222.

National trends in emergency department closures, mergers, and utilization, 2005-2015

Study objectives

To describe nationwide hospital-based emergency department (ED) closures and mergers, as well as the utilization of emergency departments and inpatient beds, over time and across varying geographic areas in the United States.

Methods

Observational analysis of the American Hospital Association (AHA) Annual Survey from 2005 to 2015. Primary outcomes were hospital-based ED closure and merger. Secondary outcomes were yearly ED visits per hospital-based ED and yearly hospital admissions per hospital bed.

Results

The total number of hospital-based EDs decreased from 4,500 in 2005 to 4,460 in 2015, with 200 closures, 138 mergers, and 160 new hospital-based EDs. While yearly ED visits per hospital-based ED exhibited a 28.6% relative increase (from 25,083 to 32,248), yearly hospital admissions per hospital bed had a 3.3% relative increase (from 45.4 to 43.9) from 2005 to 2015. The number of hospital admissions and hospital beds did not change significantly in urban areas and declined in rural areas. ED visits grew more uniformly across urban and rural areas.

Conclusions

The number of hospital-based ED closures is small when accounting for mergers, but occurs as many more patients are presenting to a stable number of EDs in larger health systems, though rural areas may differentially affected. EDs were managing accelerating patient volumes alongside stagnant inpatient bed capacity.

Letting Residents Lead: Implementing Resident Admission Triage Call Curriculum and Practice

BACKGROUND AND OBJECTIVES

Graduating residents are expected to be competent in triaging patients to appropriate resources. Before 2017, pediatric residents were not involved in admission triage decisions. In 2017, after implementing an admission triage curriculum (ATC), residents had opportunities to be involved in overnight admission calls with the emergency department (ED), which were initially supervised (joint calls), and as skills progressed, residents conducted calls and admitted patients independently. We implemented and evaluated the impact of a graduated ATC intervention bundle on pediatric resident opportunities to participate in admission triage, while monitoring resident confidence, the ED experience, and patient safety.

METHODS

We evaluated the impact of our ATC using quality improvement methodology. The primary outcome was the frequency of resident participation in joint and independent triage calls. Other measures included resident confidence, the ED clinician experience, and patient safety. Resident confidence and the ED clinician experience were rated via surveys. Safety was monitored with daytime hospitalist morning assessments and postadmission complications documented in the medical record.

RESULTS

The percent of joint calls with the hospitalist increased from 7% to 88%, and 125 patients were admitted independently. Residents reported significant increases in adequacy of triage training and confidence in 3 triage skills (P < .001) after ATC. There were no complications or safety concerns on patients admitted by residents. ED clinicians reported increased admitting process efficiency and satisfaction.

CONCLUSIONS

Our ATC intervention bundle increased the number of admission decision opportunities for pediatric residents, while increasing resident triage confidence, maintaining safety, and improving ED clinician experience.

Machine learning for developing a prediction model of hospital admission of emergency department patients: Hype or hope?

OBJECTIVE

Early identification of emergency department (ED) patients who need hospitalization is essential for quality of care and patient safety. We aimed to compare machine learning (ML) models predicting the hospitalization of ED patients and conventional regression techniques at three points in time after ED registration.

METHODS

We analyzed consecutive ED patients of three hospitals using the Netherlands Emergency Department Evaluation Database (NEED). We developed prediction models for hospitalization using an increasing number of data available at triage, ∼30 min (including vital signs) and ∼2 h (including laboratory tests) after ED registration, using ML (random forest, gradient boosted decision trees, deep neural networks) and multivariable logistic regression analysis (including spline transformations for continuous predictors). Demographics, urgency, presenting complaints, disease severity and proxies for comorbidity, and complexity were used as covariates. We compared the performance using the area under the ROC curve in independent validation sets from each hospital.

RESULTS

We included 172,104 ED patients of whom 66,782 (39 %) were hospitalized. The AUC of the multivariable logistic regression model was 0.82 (0.78-0.86) at triage, 0.84 (0.81-0.86) at ∼30 min and 0.83 (0.75-0.92) after ∼2 h. The best performing ML model over time was the gradient boosted decision trees model with an AUC of 0.84 (0.77-0.88) at triage, 0.86 (0.82-0.89) at ∼30 min and 0.86 (0.74-0.93) after ∼2 h.

CONCLUSIONS

Our study showed that machine learning models had an excellent but similar predictive performance as the logistic regression model for predicting hospital admission. In comparison to the 30-min model, the 2-h model did not show a performance improvement. After further validation, these prediction models could support management decisions by real-time feedback to medical personal.

The effect of point-of-care ultrasound on length of stay in the emergency department in children with neck swelling

BACKGROUND

Ultrasound is the imaging modality of choice in children presenting to the emergency department (ED) with soft tissue neck swelling. Point of care ultrasound (POCUS) has good accuracy when compared to comprehensive radiology department ultrasound (RADUS). POCUS could potentially improve ED length of stay (LOS) by improving efficiency. We aimed to evaluate the LOS of pediatric patients seen in ED with soft tissue neck swelling who received POCUS compared to RADUS. We determined unscheduled 30-day return visit rates in both groups as a balancing measure.

METHODS

We performed a retrospective review of the electronic medical record for our cross-sectional study of discharged patients ≤21 years of age who had a neck ultrasound performed by a credentialed POCUS physician or by the radiology department between July 2014 and January 2020. We included patients who had both POCUS and RADUS in the POCUS group. We compared median ED LOS in both groups using the Mann Whitney U test and proportion of unscheduled return visits to the ED in both groups using odds ratio and 95% CI.

RESULTS

There were 925 patients: 76 with only POCUS, 6 with POCUS and RADUS, and 843 with only RADUS performed. Median LOS in the POCUS group was 68.5 min (IQR 38.3120.3) versus 154.0 min (IQR 111.0, 211.0) in the RADUS group (p < 0.001). Return visit overall was 7.6%: 13.2% in the POCUS group versus 7.1% in the RADUS group (p = 0.07).

CONCLUSION

Pediatric patients evaluated in the ED for soft tissue neck swelling had a shorter LOS with POCUS than with RADUS without a statistically significant increase in 30-day return visits. We suggest a "POCUS First" approach to the care of these patients.

Reducing time to X-ray in emergency department ambulatory patients: a quality improvement project

This quality improvement project began when physicians and nurses at our hospital observed patients waiting for excessive periods of time for a porter to escort patients from the emergency department (ED) to medical imaging (MI). However, certain patients may not need staff escort and are able to ambulate from ED to MI by themselves. This would reduce waiting time from when the X-ray is ordered to X-ray being done, which may reduce overall ED length of stay and improve patients’ experience.

Our project aim is to decrease the time to X-ray by 50% within 6 months by having appropriate ambulatory patients walk from the ED to the X-ray department without a porter. To achieve our goal, several strategies were employed. First, brainstorm sessions were held to better understand the barriers and ways to implement the new process. Second, a patient survey was conducted to understand their thoughts on the change idea. Third, data were collected to assess the inefficiency problem on the number of times non-porter staff escorted patients due to porters being unavailable. A total of 14 PDSA (Plan-Do-Study-Act) cycles were completed between December 2018 and May 2019. A human factor specialist was consulted to examine the process for safety and optimisation of the patient journey.

In our PDSA cycles, self-ambulatory patients were compared with ambulatory patients who required an escort. An improvement was found from time to X-ray of 28 min (11 min vs 39 min). The new self-ambulatory process was implemented in June 2019 on a daily basis.