Triage Performance Across Large Language Models, ChatGPT, and Untrained Doctors in Emergency Medicine: Comparative Study.

BACKGROUND: Large language models (LLMs) have demonstrated impressive performances in various medical domains, prompting an exploration of their potential utility within the high-demand setting of emergency department (ED) triage. This study evaluated the triage proficiency of different LLMs and ChatGPT, an LLM-based chatbot, compared to professionally trained ED staff and untrained personnel. We further explored whether LLM responses could guide untrained staff in effective triage. OBJECTIVE: This study aimed to assess the efficacy of LLMs and the associated product ChatGPT in ED triage compared to personnel of varying training status and to investigate if the models' responses can enhance the triage proficiency of untrained personnel. METHODS: A total of 124 anonymized case vignettes were triaged by untrained doctors; different versions of currently available LLMs; ChatGPT; and professionally trained raters, who subsequently agreed on a consensus set according to the Manchester Triage System (MTS). The prototypical vignettes were adapted from cases at a tertiary ED in Germany. The main outcome was the level of agreement between raters' MTS level assignments, measured via quadratic-weighted Cohen κ. The extent of over- and undertriage was also determined. Notably, instances of ChatGPT were prompted using zero-shot approaches without extensive background information on the MTS. The tested LLMs included raw GPT-4, Llama 3 70B, Gemini 1.5, and Mixtral 8x7b. RESULTS: GPT-4-based ChatGPT and untrained doctors showed substantial agreement with the consensus triage of professional raters (κ=mean 0.67, SD 0.037 and κ=mean 0.68, SD 0.056, respectively), significantly exceeding the performance of GPT-3.5-based ChatGPT (κ=mean 0.54, SD 0.024; P<.001). When untrained doctors used this LLM for second-opinion triage, there was a slight but statistically insignificant performance increase (κ=mean 0.70, SD 0.047; P=.97). Other tested LLMs performed similar to or worse than GPT-4-based ChatGPT or showed odd triaging behavior with the used parameters. LLMs and ChatGPT models tended toward overtriage, whereas untrained doctors undertriaged. CONCLUSIONS: While LLMs and the LLM-based product ChatGPT do not yet match professionally trained raters, their best models' triage proficiency equals that of untrained ED doctors. In its current form, LLMs or ChatGPT thus did not demonstrate gold-standard performance in ED triage and, in the setting of this study, failed to significantly improve untrained doctors' triage when used as decision support. Notable performance enhancements in newer LLM versions over older ones hint at future improvements with further technological development and specific training.

Improving management of intravenous maintenance fluids in the emergency department of a university hospital.

OBJECTIVE: Intravenous (IV) fluid therapy is a known source of iatrogenic complications. Guideline implementation can be used to educate and guide physicians on adequate fluid management. In the emergency department (ED), a complex and interruption-driven environment, workload is high and active documentation is required to facilitate audits of fluid management quality. PATIENTS AND METHODS: Fluid management was evaluated in the ED records of adult non-critically ill patients admitted to a tertiary care center before (PRE: 1/12/2016-31/3/2017) and after (POST: 1/12/2018-31/3/2019) implementation of an educational intervention aiming to optimize IV fluid therapy in November 2018. First, the appropriateness of the 24-hour IV maintenance fluid prescription was evaluated, as prescribed by the emergency physician. Second, factors associated with appropriate prescribing were assessed, as well as the quality of fluid management documentation practice. Prescription appropriateness and documentation quality were evaluated retrospectively using a structured audit instrument and additional review by experts. RESULTS: A total of 237 patients (2.3%) were included in the PRE-intervention group and 253 patients (2.4%) in the POST-intervention group. The expert panel evaluated 214 prescriptions in 82.3% of patients (PRE: 99, POST: 115), and appropriateness increased significantly (19.2% vs. 61.2%, p=0.002). A higher odds of an appropriate IV maintenance fluid prescription was determined, attributed to the intervention (adjOR=2.580; 95% CI 1.363-4.884) and in patients having a prehospital intervention (adjOR=1.914, 95% CI 1.022-3.586). Appropriateness of fluid management documentation did not significantly improve after the implementation of the intervention (15.6% vs. 16.2%, p=0.858). CONCLUSIONS: The IV fluid prescriptions' appropriateness was significantly higher after guideline implementation. However, documentation quality of fluid management was poor in the studied ED records. Active stewardship programs are warranted to further monitor fluid management quality in the ED.

Investigating Associations Between Nonadherence to Guideline-Recommended Treatment of Pediatric Seizures and Adverse Outcomes: A Canadian Feasibility Study.

BACKGROUND: Emerging evidence suggests that nonadherence to treatment guidelines for seizures may affect patient outcomes. We examined the feasibility of conducting a larger investigation to test this hypothesis in the pediatric population. METHODS: We retrospectively reviewed charts of patients aged ≤18 years who presented with seizure to the emergency departments of two Ontario hospitals in 2019 to 2021. Patients were grouped by seizure duration (<5 minutes [n = 37], ≥5 minutes [n = 41]). We examined nonadherence to guideline-recommended treatment, adverse outcomes (hospitalization, length of stay, respiratory complications), and missing values for key variables. RESULTS: Of 78 patients, 34 (44%) did not receive guideline-recommended treatment. Nonadherence was similar in the two groups (<5 minutes: 46%; ≥5 minutes: 41%). Common deviations included administering an antiseizure medication (ASM) for seizures of less than five minutes (46%), a delay (>10 minutes) between the first and second ASM doses (50%), and use of a benzodiazepine for the third dose (45%). Hospitalizations were common in both seizure duration groups (∼90%), whereas respiratory complications were relatively rare. Time of seizure onset was missing in 51% of charts, and none contained the time of first contact with emergency services when patients were transported by ambulance. CONCLUSION: We found evidence of substantial nonadherence to guideline-recommended treatment of pediatric seizures. Medical records do not contain sufficient information to comprehensively investigate this issue. A multicenter prospective study is the most feasible option to examine the association between nonadherence to guideline-recommended treatment and patient outcomes.

Standardization of Emergency Department Clinical Note Templates: A Retrospective Analysis across an Integrated Health System.

BACKGROUND AND OBJECTIVE: Clinical documentation is essential for conveying medical decision-making, communication between providers and patients, and capturing quality, billing, and regulatory measures during emergency department (ED) visits. Growing evidence suggests the benefits of note template standardization; however, variations in documentation practices are common. The primary objective of this study is to measure the utilization and coding performance of a standardized ED note template implemented across a nine-hospital health system. METHODS: This was a retrospective study before and after the implementation of a standardized ED note template. A multi-disciplinary group consensus was built around standardized note elements, provider note workflows within the electronic health record (EHR), and how to incorporate newly required medical decision-making elements. The primary outcomes measured included the proportion of ED visits using standardized note templates, and the distribution of billing codes in the 6 months before and after implementation. RESULTS: In the preimplementation period, a total of six legacy ED note templates were being used across nine EDs, with the most used template accounting for approximately 36% of ED visits. Marked variations in documentation elements were noted across six legacy templates. After the implementation, 82% of ED visits system-wide used a single standardized note template. Following implementation, we observed a 1% increase in the proportion of ED visits coded as highest acuity and an unchanged proportion coded as second highest acuity. CONCLUSION: We observed a greater than twofold increase in the use of a standardized ED note template across a nine-hospital health system in anticipation of the new 2023 coding guidelines. The development and utilization of a standardized note template format relied heavily on multi-disciplinary stakeholder engagement to inform design that worked for varied documentation practices within the EHR. After the implementation of a standardized note template, we observed better-than-anticipated coding performance.

Use of Electronic Health Record-Based Measures to Assess Quality of Care for Pediatric Agitation.

OBJECTIVES: Acute agitation during pediatric mental health emergency department (ED) visits presents safety risks to patients and staff. We previously convened multidisciplinary stakeholders who prioritized 20 proposed quality measures for pediatric acute agitation management. Our objectives were to assess feasibility of evaluating performance on these quality measures using electronic health record (EHR) data and to examine performance variation across 3 EDs. METHODS: At a children's hospital and 2 nonchildren's hospitals, we assessed feasibility of evaluating quality measures for pediatric acute agitation management using structured EHR data elements. We retrospectively evaluated measure performance during ED visits by children 5 to 17 years old who presented for a mental health condition, received medication for agitation, or received physical restraints from July 2020 to June 2021. Bivariate and multivariable regression were used to examine measure performance by patient characteristics and hospital. RESULTS: We identified 2785 mental health ED visits, 275 visits with medication given for agitation, and 35 visits with physical restraints. Performance was feasible to measure using EHR data for 10 measures. Nine measures varied by patient characteristics, including 4.87 times higher adjusted odds (95% confidence interval 1.28-18.54) of physical restraint use among children with versus without autism spectrum disorder. Four measures varied by hospital, with physical restraint use varying from 0.5% to 3.3% of mental health ED visits across hospitals. CONCLUSIONS: Quality of care for pediatric acute agitation management was feasible to evaluate using EHR-derived quality measures. Variation in performance across patient characteristics and hospitals highlights opportunities to improve care quality.

The new ESC acute coronary syndrome guideline and its impact in the CPU and emergency department setting.

The new guideline on acute coronary syndrome (ACS) of the European Society of Cardiology (ESC) replaces two separate guidelines on ST-elevation myocardial infarction (STEMI) and non-ST-elevation (NSTE) ACS. This change of paradigm reflects the experts view that the ACS is a continuum, starting with unstable angina and ending in cardiogenic shock or cardiac arrest due to severe myocardial ischemia. Secondary, partly non-atherosclerotic-caused myocardial infarctions ("type 2") are not integrated in this concept.With respect to acute care in the setting of emergency medicine and the chest pain unit structures, the following new aspects have to be taken into account:1. New procedural approach as "think A.C.S." meaning "abnormal ECG," "clinical context," and "stable patient"2. New recommendation regarding a holistic approach for frail patients3. Revised recommendations regarding imaging and timing of invasive strategy in suspected NSTE-ACS4. Revised recommendations for antiplatelet and anticoagulant therapy in STEMI5. Revised recommendations for cardiac arrest and out-of-hospital cardiac arrest6. Revised recommendations for in-hospital management (starting in the CPU/ED) and ACS comorbid conditionsIn summary, the changes are mostly gradual and are not based on extensive new evidence, but more on focused and healthcare process-related considerations.

Adherence to Emergency Department Referral Criteria in a Direct-to-Consumer Telemedicine Center.

Background: The safety of direct-to-consumer telemedicine (TM) is closely related to red flag detection and correct referrals. The adherence to referral criteria from current guidelines is not well quantified. Objective: To analyze the emergency department (ED) referral rate and adherence to referral guidelines in TM encounters of acutely ill patients calling a center that adopts stewardship protocols. Methods: This is a retrospective observational unicentric study, between March 2020 and March 2022, with patients who spontaneously sought direct-to-consumer urgent virtual medical assistance. A video-based teleconsultation was provided immediately after connection. Physicians managed situations according to their clinical judgment. Current guidelines, containing specific guidance for referral if red flags were identified, were available for consultation. Physicians' semiannual performance feedback was carried out. We analyzed the patterns for referral to immediate face-to-face medical evaluation and the agreement degree with the institutional guidelines. Results: A total of 232,197 patients were available, and 14,051 (6.05%) patients were referred to ED. A total of 8,829 (68.4%) referrals were based in specific guidelines according to the International Classification of Diseases hypothesis, and 8,708 (98.6%) were justified according to guidelines. Diarrhea had the highest guidelines' adherence to referral (97.6%), followed by COVID-19 (90%), headache (84.2%), and conjunctivitis (78.8%). Policies did not support 5,222 (31.6%) referrals, though 5,100 (97.6%) of these were justified according to the doctor's clinical judgment. Conclusion: TM doctors' assessment of acutely ill patients has high rates of adherence to guidelines regarding referral. Stewardship protocol adoption provides high rates of red flag description, even in the referral of nonpolicy diseases.

Improving Outcomes in Patients Sent to the Emergency Department from Outpatient Providers: A Receiver-Driven Handoff Process Improvement.

BACKGROUND: Outpatient providers refer to emergency departments (EDs) due to findings requiring assessment beyond existing capabilities. However, poor communication surrounding these transitions may hinder safety and timeliness of emergency care. Receiver-driven handoff (RDH) is a process that helps ensure that all pertinent information is shared. This quality improvement project aimed to (1) improve knowledge of RDH, (2) increase satisfaction and perceptions surrounding RDH, (3) modify behaviors in relation to RDH, and (4) decrease referred patients leaving without being seen (LWBS). METHODS: The Iowa Model and Implementation Framework guided this evidence-based quality improvement project. A multidisciplinary team developed and implemented a standardized RDH process consisting of screening to determine whether a patient was referred to the ED, review of electronic health record (EHR), and use of EHR documentation. Process measures were collected via questionnaire pre- and postimplementation and were analyzed quantitatively. Outcome measures were trended by a statistical process control p-chart, which was developed to demonstrate changes in the percentage of patients who were referred to the ED from the outpatient setting and LWBS. RESULTS: The average response for the question "How satisfied are you with the handoff of patient information from referring clinic providers to the ED?" increased from 1.51 preintervention to 2.04 postintervention (p = 0.005). Respondents rated the information received during handoff higher postintervention (2.12 vs. 2.52, p = 0.04). Compliance with screening for referral to the ED was 84.0%. The proportion of patients LWBS after referral decreased by 6.2 percentage points (p < 0.001). CONCLUSION: Using RDH in conjunction with a standardized triage screening may improve quality of information shared during this vulnerable transition and may assist in reduction of referred patients LWBS. The RDH process should be adapted into everyday workflow to ensure sustainability and effectiveness.

Improving outcomes in emergency general surgery: Construct of a collaborative quality initiative.

BACKGROUND: Emergency general surgery conditions are common, costly, and highly morbid. The proportion of excess morbidity due to variation in health systems and processes of care is poorly understood. We constructed a collaborative quality initiative for emergency general surgery to investigate the emergency general surgery care provided and guide process improvements. METHODS: We collected data at 10 hospitals from July 2019 to December 2022. Five cohorts were defined: acute appendicitis, acute gallbladder disease, small bowel obstruction, emergency laparotomy, and overall aggregate. Processes and inpatient outcomes investigated included operative versus nonoperative management, mortality, morbidity (mortality and/or complication), readmissions, and length of stay. Multivariable risk adjustment accounted for variations in demographic, comorbid, anatomic, and disease traits. RESULTS: Of the 19,956 emergency general surgery patients, 56.8% were female and 82.8% were White, and the mean (SD) age was 53.3 (20.8) years. After accounting for patient and disease factors, the adjusted aggregate mortality rate was 3.5% (95% confidence interval [CI], 3.2-3.7), morbidity rate was 27.6% (95% CI, 27.0-28.3), and the readmission rate was 15.1% (95% CI, 14.6-15.6). Operative management varied between hospitals from 70.9% to 96.9% for acute appendicitis and 19.8% to 79.4% for small bowel obstruction. Significant differences in outcomes between hospitals were observed with high- and low-outlier performers identified after risk adjustment in the overall cohort for mortality, morbidity, and readmissions. The use of a Gastrografin challenge in patients with a small bowel obstruction ranged from 10.7% to 61.4% of patients. In patients who underwent initial nonoperative management of acute cholecystitis, 51.5% had a cholecystostomy tube placed. The cholecystostomy tube placement rate ranged from 23.5% to 62.1% across hospitals. CONCLUSION: A multihospital emergency general surgery collaborative reveals high morbidity with substantial variability in processes and outcomes among hospitals. A targeted collaborative quality improvement effort can identify outliers in emergency general surgery care and may provide a mechanism to optimize outcomes. LEVEL OF EVIDENCE: Therapeutic/Care Management; Level III.

Door-in-Door-out Times for Interhospital Transfer of Patients With Stroke.

Importance: Treatments for time-sensitive acute stroke are not available at every hospital, often requiring interhospital transfer. Current guidelines recommend hospitals achieve a door-in-door-out time of no more than 120 minutes at the transferring emergency department (ED). Objective: To evaluate door-in-door-out times for acute stroke transfers in the American Heart Association Get With The Guidelines-Stroke registry and to identify patient and hospital factors associated with door-in-door-out times. Design, Setting, and Participants: US registry-based, retrospective study of patients with ischemic or hemorrhagic stroke from January 2019 through December 2021 who were transferred from the ED at registry-affiliated hospitals to other acute care hospitals. Exposure: Patient- and hospital-level characteristics. Main Outcomes and Measures: The primary outcome was the door-in-door-out time (time of transfer out minus time of arrival to the transferring ED) as a continuous variable and a categorical variable (≤120 minutes, >120 minutes). Generalized estimating equation (GEE) regression models were used to identify patient and hospital-level characteristics associated with door-in-door-out time overall and in subgroups of patients with hemorrhagic stroke, acute ischemic stroke eligible for endovascular therapy, and acute ischemic stroke transferred for reasons other than endovascular therapy. Results: Among 108 913 patients (mean [SD] age, 66.7 [15.2] years; 71.7% non-Hispanic White; 50.6% male) transferred from 1925 hospitals, 67 235 had acute ischemic stroke and 41 678 had hemorrhagic stroke. Overall, the median door-in-door-out time was 174 minutes (IQR, 116-276 minutes): 29 741 patients (27.3%) had a door-in-door-out time of 120 minutes or less. The factors significantly associated with longer median times were age 80 years or older (vs 18-59 years; 14.9 minutes, 95% CI, 12.3 to 17.5 minutes), female sex (5.2 minutes; 95% CI, 3.6 to 6.9 minutes), non-Hispanic Black vs non-Hispanic White (8.2 minutes, 95% CI, 5.7 to 10.8 minutes), and Hispanic ethnicity vs non-Hispanic White (5.4 minutes, 95% CI, 1.8 to 9.0 minutes). The following were significantly associated with shorter median door-in-door-out time: emergency medical services prenotification (-20.1 minutes; 95% CI, -22.1 to -18.1 minutes), National Institutes of Health Stroke Scale (NIHSS) score exceeding 12 vs a score of 0 to 1 (-66.7 minutes; 95% CI, -68.7 to -64.7 minutes), and patients with acute ischemic stroke eligible for endovascular therapy vs the hemorrhagic stroke subgroup (-16.8 minutes; 95% CI, -21.0 to -12.7 minutes). Among patients with acute ischemic stroke eligible for endovascular therapy, female sex, Black race, and Hispanic ethnicity were associated with a significantly higher door-in-door-out time, whereas emergency medical services prenotification, intravenous thrombolysis, and a higher NIHSS score were associated with significantly lower door-in-door-out times. Conclusions and Relevance: In this US registry-based study of interhospital transfer for acute stroke, the median door-in-door-out time was 174 minutes, which is longer than current recommendations for acute stroke transfer. Disparities and modifiable health system factors associated with longer door-in-door-out times are suitable targets for quality improvement initiatives.

Profiles of quality of outpatient care use, associated sociodemographic and clinical characteristics, and adverse outcomes among patients with substance-related disorders.

BACKGROUND: This study identified patient profiles in terms of their quality of outpatient care use, associated sociodemographic and clinical characteristics, and adverse outcomes based on frequent emergency department (ED) use, hospitalization, and death from medical causes. METHODS: A cohort of 18,215 patients with substance-related disorders (SRD) recruited in addiction treatment centers was investigated using Quebec (Canada) health administrative databases. A latent class analysis was produced, identifying three profiles of quality of outpatient care use, while multinomial and logistic regressions tested associations with patient characteristics and adverse outcomes, respectively. RESULTS: Profile 1 patients (47% of the sample), labeled "Low outpatient service users", received low quality of care. They were mainly younger, materially and socially deprived men, some with a criminal history. They had more recent SRD, mainly polysubstance, and less mental disorders (MD) and chronic physical illnesses than other Profiles. Profile 2 patients (36%), labeled "Moderate outpatient service users", received high continuity and intensity of care by general practitioners (GP), while the diversity and regularity in their overall quality of outpatient service was moderate. Compared with Profile 1, they  were older, less likely to be unemployed or to live in semi-urban areas, and most had common MD and chronic physical illnesses. Profile 3 patients (17%), labeled "High outpatient service users", received more intensive psychiatric care and higher quality of outpatient care than other Profiles. Most Profile 3 patients lived alone or were single parents, and fewer lived in rural areas or had a history of homelessness, versus Profile 1 patients. They were strongly affected by MD, mostly serious MD and personality disorders. Compared with Profile 1, Profile 3 had more frequent ED use and hospitalizations, followed by Profile 2. No differences in death rates emerged among the profiles. CONCLUSIONS: Frequent ED use and hospitalization were strongly related to patient clinical and sociodemographic profiles, and the quality of outpatient services received to the severity of their conditions. Outreach strategies more responsive to patient needs may include motivational interventions and prevention of risky behaviors for Profile 1 patients, collaborative GP-psychiatrist care for Profile 2 patients, and GP care and intensive specialized treatment for Profile 3 patients.

Knowledge, attitude, and practice of health professionals working in emergency units towards disaster and emergency preparedness in South Gondar Zone hospitals, Ethiopia, 2020.

Introduction: catastrophe is a thoughtful community's well-being problem nowadays. Tragedies of any kind can strike at any time and have claimed many lives. Because, the emergency unit is at the frontline of disaster/emergency response system and helps as initial point to the most proper care of causalities, health professionals who are working in this area are the first caregivers, main players, and upfront role in calamity responses after pre-hospital medical services to disaster victims. The aim of this study was to assess emergency unit health professionals´ knowledge, attitude, practice, and related factors towards disasters and emergency preparedness at hospitals in the South Gondar Zone, Ethiopia, 2020. Methods: institution-based cross-sectional study with the census method was conducted at South Gondar Zone hospitals. All health professionals working in emergency units of South Gondar Zone hospitals were taken as a sample. A structured self-administered questionnaire was used to collect data. EPI-data version 4.2 and SPSS version 25 were used to enter and analyze data, respectively. The result was presented by narration, tables, and charts. Binary logistic regression was employed to determine the relations between dependent and independent variables. Results: the majority of the respondents (58.3%) were male. Regarding their profession, 52.2% were nurses, followed by physicians, 18.5%, while the rest were others. The mean age of the respondents was 29.48 ± 6.34 years. A substantial proportion (58.9%) of the study participants didn´t know whether their hospitals had a disaster management plan or not. In general, fifty-one-point seven percent´s (51.7%) of the study participants had poor knowledge toward disaster/emergency preparedness. Concerning their attitude, 55.0% had a negative attitude toward disaster preparedness. Regarding their levels of practice, 67.5% had inadequate practice disaster/emergency preparedness. Age category and profession of the respondents had a significant effect on the knowledge and attitude of respondents at P-value 0.05. Conclusion: more than half of the study participants had poor knowledge, negative attitudes, and inadequate practice about disaster/emergency preparedness.

Oilcloth sessions as an implementation strategy: a qualitative study in Denmark.

BACKGROUND: The aim of this study was to explore healthcare professionals, managers, and other key employees' experiences of oilcloth sessions as a strategy when implementing new emergency departments in Denmark, based on their participations in these sessions. The study addresses the importance of securing alignment in implementation strategies. Too often, this does not get enough attention in the literature and in practice. In this study, alignment among components was achieved in an educational implementation strategy called oilcloth sessions. METHODS: The study is based on participants' observations of 13 oilcloth sessions and follow-up via 53 semi-structured interviews with the board of directors, managers, and key employees from the present emergency department and different specialty departments. Data were analysed deductively using Biggs and Tang's model of didactic alignment. RESULTS: The analysis showed the complexity of challenges when using oilcloth sessions as a strategy when implementing a new emergency department described in terms of three phases and nine main themes (a-i): the preparation phase: (a) preparing individually and collectively, (b) objectives, (c) involving participants, (d) selecting cases; the execution phase: (e) using materials, (f) facilitating the sessions, (g) temporal structures; evaluation: (h) following up on the sessions, (i) adapting to the context. CONCLUSIONS: This study shows that it is important to ensure alignment among elements in implementation strategies. Thus, oilcloth sessions with high alignment are useful if the challenges experienced are to be overcome and the strategy will be experienced as a useful way to support the implementation of a new emergency department from the participants' point of view. Bigg and Tang's didactic model is useful as an analytical framework to ensure alignment in implementation strategies in general.

Operations management solutions to improve ED patient flows: evidence from the Italian NHS.

BACKGROUND: Overcrowding occurs when the identified need for emergency services outweighs the available resources in the emergency department (ED). Literature shows that ED overcrowding impacts the overall quality of the entire hospital production system, as confirmed by the recent COVID-19 pandemic. This study aims to identify the most relevant variables that cause ED overcrowding using the input-process-output model with the aim of providing managers and policy makers with useful hints for how to effectively redesign ED operations. METHODS: A mixed-method approach is used, blending qualitative inquiry with quantitative investigation in order to: i) identifying and operationalizing the main components of the model that can be addressed by hospital operation management teams and ii) testing and measuring how these components can influence ED LOS. RESULTS: With a dashboard of indicators developed following the input-process-output model, the analysis identifies the most significant variables that have an impact on ED overcrowding: the type (age and complexity) and volume of patients (input), the actual ED structural capacity (in terms of both people and technology) and the ED physician-to-nurse ratio (process), and the hospital discharging process (output). CONCLUSIONS: The present paper represents an original contribution regarding two different aspects. First, this study combines different research methodologies with the aim of capturing relevant information that by relying on just one research method, may otherwise be missed. Second, this study adopts a hospitalwide approach, adding to our understanding of ED overcrowding, which has thus far focused mainly on single aspects of ED operations.

[The COVID-19 pandemic's impacts on adverse events associated with care]. / Impact de la pandémie COVID-19 sur les évènements indésirables associés aux soins.

Since early 2020, the onset of the COVID-19 pandemic, physicians have continued to report adverse events associated with care. Patients also continued to participate in the hospital satisfaction surveys. To date, no study in France has measured the impact of the pandemic on adverse events and patient satisfaction. We looked at the characteristics of these adverse events in relation to the pandemic and put patients' feelings into perspective. A qualitative and observational retrospective study of the REX and MCO48 databases was carried out. The quantitative study of the REX database was supplemented by a qualitative analysis of the declarations. The adverse events more often affects middle-aged men aged 60 years, while deaths occur in older patients with more complex pathologies and more urgent management. The nature of these events is different depending on the reporting period: Those reported in the first wave are more urgent, occur less frequently in the operating room than in the emergency room, and are considered less preventable than those reported in the second wave. The latter are more similar to the events that usually occur. The implementation of effective barriers, particularly within the teams, has made it possible to reduce the impact of the second wave on the occurrence of these events, the role of communication seems essential. The overall patient satisfaction score as well as those for medical and paramedical care has increased, which may reflect patient solidarity with caregivers. The attitude of active resilience on the part of all actors has been a major element in risk management during this crisis and it is essential to capitalize on these collaborative processes for the future.

Machine learning-based prediction of critical illness in children visiting the emergency department.

OBJECTIVES: Triage is an essential emergency department (ED) process designed to provide timely management depending on acuity and severity; however, the process may be inconsistent with clinical and hospitalization outcomes. Therefore, studies have attempted to augment this process with machine learning models, showing advantages in predicting critical conditions and hospitalization outcomes. The aim of this study was to utilize nationwide registry data to develop a machine learning-based classification model to predict the clinical course of pediatric ED visits. METHODS: This cross-sectional observational study used data from the National Emergency Department Information System on emergency visits of children under 15 years of age from January 1, 2016, to December 31, 2017. The primary and secondary outcomes were to identify critically ill children and predict hospitalization from triage data, respectively. We developed and tested a random forest model with the under sampled dataset and validated the model using the entire dataset. We compared the model's performance with that of the conventional triage system. RESULTS: A total of 2,621,710 children were eligible for the analysis and included 12,951 (0.5%) critical outcomes and 303,808 (11.6%) hospitalizations. After validation, the area under the receiver operating characteristic curve was 0.991 (95% confidence interval [CI] 0.991-0.992) for critical outcomes and 0.943 (95% CI 0.943-0.944) for hospitalization, which were higher than those of the conventional triage system. CONCLUSIONS: The machine learning-based model using structured triage data from a nationwide database can effectively predict critical illness and hospitalizations among children visiting the ED.

Prevalence, Characteristics, and Outcomes of Emergency Department Discharge Among Patients With Sepsis.

Importance: Sepsis guidelines and research have focused on patients with sepsis who are admitted to the hospital, but the scope and implications of sepsis that is managed in an outpatient setting are largely unknown. Objective: To identify the prevalence, risk factors, practice variation, and outcomes for discharge to outpatient management of sepsis among patients presenting to the emergency department (ED). Design, Setting, and Participants: This cohort study was conducted at the EDs of 4 Utah hospitals, and data extraction and analysis were performed from 2017 to 2021. Participants were adult ED patients who presented to a participating ED from July 1, 2013, to December 31, 2016, and met sepsis criteria before departing the ED alive and not receiving hospice care. Exposures: Patient demographic and clinical characteristics, health system parameters, and ED attending physician. Main Outcomes and Measures: Information on ED disposition was obtained from electronic medical records, and 30-day mortality data were acquired from Utah state death records and the US Social Security Death Index. Factors associated with ED discharge rather than hospital admission were identified using penalized logistic regression. Variation in ED discharge rates between physicians was estimated after adjustment for potential confounders using generalized linear mixed models. Inverse probability of treatment weighting was used in the primary analysis to assess the noninferiority of outpatient management for 30-day mortality (noninferiority margin of 1.5%) while adjusting for multiple potential confounders. Results: Among 12 333 ED patients with sepsis (median [IQR] age, 62 [47-76] years; 7017 women [56.9%]) who were analyzed in the study, 1985 (16.1%) were discharged from the ED. After penalized regression, factors associated with ED discharge included age (adjusted odds ratio [aOR], 0.90 per 10-y increase; 95% CI, 0.87-0.93), arrival to ED by ambulance (aOR, 0.61; 95% CI, 0.52-0.71), organ failure severity (aOR, 0.58 per 1-point increase in the Sequential Organ Failure Assessment score; 95% CI, 0.54-0.60), and urinary tract (aOR, 4.56 [95% CI, 3.91-5.31] vs pneumonia), intra-abdominal (aOR, 0.51 [95% CI, 0.39-0.65] vs pneumonia), skin (aOR, 1.40 [95% CI, 1.14-1.72] vs pneumonia) or other source of infection (aOR, 1.67 [95% CI, 1.40-1.97] vs pneumonia). Among 89 ED attending physicians, adjusted ED discharge probability varied significantly (likelihood ratio test, P < .001), ranging from 8% to 40% for an average patient. The unadjusted 30-day mortality was lower in discharged patients than admitted patients (0.9% vs 8.3%; P < .001), and their adjusted 30-day mortality was noninferior (propensity-adjusted odds ratio, 0.21 [95% CI, 0.09-0.48]; adjusted risk difference, 5.8% [95% CI, 5.1%-6.5%]; P < .001). Alternative confounder adjustment strategies yielded odds ratios that ranged from 0.21 to 0.42. Conclusions and Relevance: In this cohort study, discharge to outpatient treatment of patients who met sepsis criteria in the ED was more common than previously recognized and varied substantially between ED physicians, but it was not associated with higher mortality compared with hospital admission. Systematic, evidence-based strategies to optimize the triage of ED patients with sepsis are needed.

First Nations status and emergency department triage scores in Alberta: a retrospective cohort study.

BACKGROUND: Previous studies have found that race is associated with emergency department triage scores, raising concerns about potential health care inequity. As part of a project on quality of care for First Nations people in Alberta, we sought to understand the relation between First Nations status and triage scores. METHODS: We conducted a population-based retrospective cohort study of health administrative data from April 2012 to March 2017 to evaluate acuity of triage scores, categorized as a binary outcome of higher or lower acuity score. We developed multivariable multilevel logistic mixed-effects regression models using the levels of emergency department visit, patient (for patients with multiple visits) and facility. We further evaluated the triage of visits related to 5 disease categories and 5 specific diagnoses to better compare triage outcomes of First Nations and non-First Nations patients. RESULTS: First Nations status was associated with lower odds of receiving higher acuity triage scores (odds ratio [OR] 0.93, 95% confidence interval [CI] 0.92-0.94) compared with non-First Nations patients in adjusted models. First Nations patients had lower odds of acute triage for all 5 disease categories and for 3 of 5 diagnoses, including long bone fractures (OR 0.82, 95% CI 0.76-0.88), acute upper respiratory infection (OR 0.90, 95% CI 0.84-0.98) and anxiety disorder (OR 0.67, 95% CI 0.60-0.74). INTERPRETATION: First Nations status was associated with lower odds of higher acuity triage scores across a number of conditions and diagnoses. This may reflect systemic racism, stereotyping and potentially other factors that affected triage assessments.