Development and validation of a machine learning framework for improved resource allocation in the emergency department.

OBJECTIVE: The Emergency Severity Index (ESI) is the most commonly used system in over 70% of all U.S. emergency departments (ED) that uses predicted resource utilization as a means to triage [1], Mistriage, which includes both undertriage and overtriage has been a persistent issue, affecting 32.2% of total ED visits [2]. Our goal is to develop a machine learning framework that predicts patients' resource needs, thereby improving resource allocation during triage. METHODS: This retrospective study analyzed ED visits from the Medical Information Mart for Intensive Care IV, dividing the data into training (80%) and testing (20%) cohorts. We utilized data available during triage, including patient vital signs, age, gender, mode of arrival, medication history, and chief complaint. Azure AutoML was used to create different machine learning models trained to predict the 144 target columns including laboratory panels and imaging modalities as well as medications required during patients' ED visits. The 144 models' performance was evaluated using the area under the receiver operating characteristic curve (AUROC), F1 score, accuracy, precision and recall. RESULTS: A total of 391,472 ED visits were analyzed. 144 Voting ensemble models were created for each target. All frameworks achieved on average an AUC score of 0.82 and accuracy of 0.76. We gathered the feature importance for each target and observed that 'chief complaint', among others, had a high aggregate feature importance across different targets. CONCLUSION: This study shows the high accuracy in predicting resource needs for patients in the ED using a machine learning model. This can greatly improve patient flow and resource allocation in already resource limited emergency departments.

CT utilization in evaluation of skin and soft tissue extremity infections in the ED: Retrospective cohort study.

OBJECTIVE: Skin and soft tissue infections (SSTI) are commonly diagnosed in the emergency department (ED). While most SSTI are diagnosed with patient history and physical exam alone, ED clinicians may order CT imaging when they suspect more serious or complicated infections. Patients who inject drugs are thought to be at higher risk for complications from SSTI and may undergo CT imaging more frequently. The objective of this study is to characterize CT utilization when evaluating for SSTI in ED patients particularly in patients with intravenous drug use (IVDU), the frequency of significant and actionable findings from CT imaging, and its impact on subsequent management and ED operations. METHODS: We performed a retrospective analysis of encounters involving a diagnosis of SSTI in seven EDs across an integrated health system between October 2019 and October 2021. Descriptive statistics were used to assess overall trends, compare CT utilization frequencies, actionable imaging findings, and surgical intervention between patients who inject drugs and those who do not. Multivariable logistic regression was used to analyze patient factors associated with higher likelihood of CT imaging. RESULTS: There were 4833 ED encounters with an ICD-10 diagnosis of SSTI during the study period, of which 6% involved a documented history of IVDU and 30% resulted in admission. 7% (315/4833) of patients received CT imaging, and 22% (70/315) of CTs demonstrated evidence of possible deep space or necrotizing infections. Patients with history of IVDU were more likely than patients without IVDU to receive a CT scan (18% vs 6%), have a CT scan with findings suspicious for deep-space or necrotizing infection (4% vs 1%), and undergo surgical drainage in the operating room within 48 h of arrival (5% vs 2%). Male sex, abnormal vital signs, and history of IVDU were each associated with higher likelihood of CT utilization. Encounters involving CT scans had longer median times to ED disposition than those without CT scans, regardless of whether these encounters resulted in admission (9.0 vs 5.5 h), ED observation (5.5 vs 4.1 h), or discharge (6.8 vs 2.9 h). DISCUSSION: ED clinicians ordered CT scans in 7% of encounters when evaluating for SSTI, most frequently in patients with abnormal vital signs or a history of IV drug use. Patients with a history of IVDU had higher rates of CT findings suspicious for deep space infections or necrotizing infections and higher rates of incision and drainage procedures in the OR. While CT scans significantly extended time spent in the ED for patients, this appeared justified by the high rate of actionable findings found on imaging, particularly for patients with a history of IVDU.

Predicting daily emergency department visits using machine learning could increase accuracy.

OBJECTIVE: Administrators and clinicians alike have attempted to predict emergency department visits for many years. The ability to predict or "forecast" ED visit volume can allow for more efficient resource allocation, including up-staffing or down-staffing, changing OR schedules, and predicting the need for significant resources. The goal of this study is to examine combinations of variables via machine learning to increase prediction accuracy and determine the factors that are most predictive of overall ED visits. As compared to a simple univariate time series model, we hypothesize that machine learning models will predict St. Joseph Mercy Ann Arbor's patient visit load for the emergency department (ED) with higher accuracy than a simple univariate time series model. METHODS: Univariate time series models for daily ED visits, including ARIMA, Exponential Smoothing (ETS), and Facebook Inc.'s prophet algorithm were estimated as a baseline comparison. Machine learning models, including random forests and gradient boosted machines (GBM), were trained using data from 2017 to 2018. After final models were created, they were applied to the 2019 data to determine how well these models predicted actual ED patient volumes in data not utilized during the model fitting process. The accuracy of the machine learning and time series models were assessed based on out-of-sample predictive accuracy, compared using root mean squared error (RMSE). RESULTS: Using root mean squared error (RMSE) to assess out-of-sample predictive accuracy of the models, the results showed that the random forest model was the most accurate at predicting daily ED visits in the 2019 test set, followed by the GBM model. These performed only slightly better than the simple exponential smoothing model predictions. The ARIMA model performed poorly in comparison. The day of the week (likely capturing differences between weekdays and weekends) was found to be the most important predictor of patient volumes. Weather-related features such as maximum temperature and SFC pressure appeared to capture some of the seasonality trends related to changes in patient volumes. CONCLUSIONS: Machine learning models perform better at predicting daily patient volumes as compared to simple univariate time series models, though not by a substantial amount. Further research can help confirm these limited initial results. Gathering more training data and additional feature engineering could also be beneficial to training the models and potentially improving predictive accuracy.

Impact of iodinated contrast allergies on emergency department operations.

OBJECTIVES: Adverse reactions to intravenous (IV) iodinated contrast media are classified by the American College of Radiology (ACR) Manual on Contrast Media as either allergic-like (ALR) or physiologic (PR). Premedication may be beneficial for patients who have prior documented mild or moderate ALR. We sought to perform a retrospective analysis of patients who received computed tomography (CT) imaging in our emergency department (ED) to establish whether listing of an iodinated contrast media allergy results in a delay in care, increases the use of non-contrast studies, and to quantify the incidence of listing iodinated contrast allergies which do not necessitate premedication. METHODS: We performed a retrospective analysis of CT scans performed in our academic medical center ED during a 6-month period. There were 12,737 unique patients of whom 454 patients had a listed iodinated contrast allergy. Of these, 106 received IV contrast and were categorized as to whether premedication was necessary. Descriptive statistics were used to evaluate patient demographics, clinical characteristics, and operational outcomes. A multivariate linear regression model was used to predict time from order to start (OTS time) of CT imaging while controlling for co-variates. RESULTS: Non-allergic patients underwent contrast-enhanced CT imaging at a significantly higher rate than allergic patients (45.9% vs. 23.3%, p < 0.01). The OTS time for allergic patients who underwent contrast-enhanced CT imaging was 360 min and significantly longer than the OTS time for non-allergic patients who underwent contrast-enhanced CT imaging (118 min, p < 0.001). Of the 106 allergic patients who underwent contrast-enhanced CT imaging, 27 (25.5%) did not meet ACR criteria for necessitating premedication. The average OTS time for these 27 patients was 296 min, significantly longer than the OTS for non-allergic patients (118 min, p < 0.01) and did not differ from the OTS time for the 79 patients who did meet premedication criteria (382 min, p = 0.23). A multivariate linear regression showed that OTS time was significantly longer if a contrast allergy was present (p < 0.001). CONCLUSION: A chart-documented iodinated contrast allergy resulted in a significant increase in time to obtain a contrast-enhanced CT study. This delay persisted among patients who did not meet ACR criteria for premedication. Appropriately deferring premedication could potentially reduce the ED length-of-stay by over 4 h for these patients.

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 emergency department boarding times on hospital length of stay for patients with psychiatric illness.

BACKGROUND: Extended periods awaiting an inpatient bed in the emergency department (ED) may exacerbate the state of patients with acute psychiatric illness, increasing the time it takes to stabilise their acute problem in hospital. Therefore, we assessed the association between boarding time and hospital length of stay for psychiatric patients. METHODS: ED clinical records were linked to inpatient administrative records for all patients with a primary psychiatric diagnosis admitted to a Calgary, Alberta hospital between April 2014 and March 2018. The primary exposure was boarding time (admission decision to inpatient bed transfer), and primary outcome was inpatient length of stay. Confounders for this relationship, including indicators of illness severity, were selected a priori then the association was assessed using hierarchical Bayesian Poisson regression, which accounts for repeat observations of the same patient and differences between hospital sites. Changes in length of stay were measured using a rate ratio (ie, expected change in length of stay for each 1 hour increase in boarding time). RESULTS: A total of 19 212 admissions (14 261 unique patients) were included in the analysis. The average boarding time was 14 hours (range: 0-186 hours). Patients who were boarded for greater than 14 hours more frequently required a high-observation bed (14% vs 3.5%), received an antipsychotic (44% vs 14%) or received sedation (55% vs 33%) while in the ED. The probability that boarding time increased hospital length of stay (rate ratio: >1) was 92%, with a median increase for a patient boarded for 24 hours of 0.01 days. CONCLUSION: Boarding in the ED was associated with a high probability of increasing the hospital length of stay for psychiatric patients; however, the absolute increase is minimal. Although slight, this signal for longer length of stay may be a sign of increased morbidity for psychiatric patients held in the ED.

Emergency medicine patient wait time multivariable prediction models: a multicentre derivation and validation study.

OBJECTIVE: Patients, families and community members would like emergency department wait time visibility. This would improve patient journeys through emergency medicine. The study objective was to derive, internally and externally validate machine learning models to predict emergency patient wait times that are applicable to a wide variety of emergency departments. METHODS: Twelve emergency departments provided 3 years of retrospective administrative data from Australia (2017-2019). Descriptive and exploratory analyses were undertaken on the datasets. Statistical and machine learning models were developed to predict wait times at each site and were internally and externally validated. Model performance was tested on COVID-19 period data (January to June 2020). RESULTS: There were 1 930 609 patient episodes analysed and median site wait times varied from 24 to 54 min. Individual site model prediction median absolute errors varied from±22.6 min (95% CI 22.4 to 22.9) to ±44.0 min (95% CI 43.4 to 44.4). Global model prediction median absolute errors varied from ±33.9 min (95% CI 33.4 to 34.0) to ±43.8 min (95% CI 43.7 to 43.9). Random forest and linear regression models performed the best, rolling average models underestimated wait times. Important variables were triage category, last-k patient average wait time and arrival time. Wait time prediction models are not transferable across hospitals. Models performed well during the COVID-19 lockdown period. CONCLUSIONS: Electronic emergency demographic and flow information can be used to approximate emergency patient wait times. A general model is less accurate if applied without site-specific factors.

The impact of COVID-19 on an Irish Emergency Department (ED): a cross-sectional study exploring the factors influencing ED utilisation prior to and during the pandemic from the patient perspective.

BACKGROUND: The collateral damage of SARS-CoV-2 is a serious concern in the Emergency Medicine (EM) community, specifically in relation to delayed care increasing morbidity and mortality in attendances unrelated to COVID-19. The objectives of this study are to describe the profile of patients attending an Irish ED prior to, and during the pandemic, and to investigate the factors influencing ED utilisation in this cohort. METHODS: This was a cross-sectional study with recruitment at three time-points prior to the onset of COVID-19 in December 2019 (n = 47) and February 2020 (n = 57) and post-Lockdown 1 in July 2020 (n = 70). At each time-point all adults presenting over a 24 h period were eligible for inclusion. Clinical data were collected via electronic records and a questionnaire provided information on demographics, healthcare utilisation, service awareness and factors influencing the decision to attend the ED. Data analysis was performed in SPSS and included descriptive and inferential statistics. RESULTS: The demographic and clinical profile of patients across time-points was comparable in terms of age (p = 0.904), gender (p = 0.584) and presenting complaint (p = 0.556). Median length of stay in the ED decreased from 7.25 h (IQR 4.18-11.22) in February to 3.86 h (IQR 0.41-9.14) in July (p ≤ 0.005) and differences were observed in disposition (p ≤ 0.001). COVID-19 influenced decision to attend the ED for 31% of patients with 9% delaying presentation. Post-lockdown, patients were less likely to attend the ED for reassurance (p ≤ 0.005), for a second opinion (p ≤ 0.005) or to see a specialist (p ≤ 0.05). CONCLUSIONS: Demographic and clinical presentations of ED patients prior to the first COVID-19 lockdown and during the reopening phase were comparable, however, COVID-19 significantly impacted health-seeking behaviour and operational metrics in the ED at this phase of the pandemic. These findings provide useful information for hospitals with regard to pandemic preparedness and also have wider implications for planning of future health service delivery.

The International Federation for Emergency Medicine report on emergency department crowding and access block: A brief summary.

OBJECTIVE: To develop comprehensive guidance that captures international impacts, causes and solutions related to emergency department (ED) crowding and access block. METHODS: Emergency physicians representing 15 countries from all International Federation of Emergency Medicine (IFEM) regions composed the Task Force. Monthly meetings were held via video-conferencing software to achieve consensus for report content. The report was submitted and approved by the IFEM Board on June 1, 2020. RESULTS: A total of 14 topic dossiers, each relating to an aspect of ED crowding, were researched and completed collaboratively by members of the Task Force. CONCLUSIONS: The IFEM report is a comprehensive document intended to be used in whole or by section to inform and address aspects of ED crowding and access block. Overall, ED crowding is a multifactorial issue requiring systems-wide solutions applied at local, regional, and national levels. Access block is the predominant contributor of ED crowding in most parts of the world.

'The incident triage area': a response to the COVID-19 pandemic in the Bristol Royal Infirmary.

With the onset of the COVID-19 pandemic, hospitals nationwide have been presented with a number of potential challenges, including possible increased volume of patient attendances, acuity of illness and potential for patients to present with an infection that requires isolation. At the Bristol Royal Infirmary, an innercity teaching hospital that manages patients aged 16 and over, we present our response to these projected changes in ED attendances, with the initiation of the incident triage area (ITA). The ITA is a triage station situated outside the ED and staffed by a senior clinician, healthcare assistant and patient flow coordinator. It receives patients presenting as walk-in or via ambulance, and on their arrival aims to establish their risk of COVID-19 and their acuity of illness. This allows for triage of the patient to one of the four zones of the hospital, as well as providing clinical guidance on any initial interventions that patients may require. The benefits of the ITA are that it enables an early senior review of patients to establish their acuity of illness and initiate time-critical medical intervention as required. In addition, patients are immediately cohorted to zones within the hospital based on their infection risk, thereby reducing patient footfall throughout the hospital. Its aim is to reduce the spread of infection, by efficiently triaging and streaming patients who present to the hospital prior to them entering clinical areas, while maintaining patient safety and flow through the ED and initiating rapid management of acutely unwell patients.

Emergency medicine response to the COVID-19 pandemic in England: a phenomenological study.

BACKGROUND: The COVID-19 pandemic has stretched EDs globally, with many regions in England challenged by the number of COVID-19 presentations. In order to rapidly share learning to inform future practice, we undertook a thematic review of ED operational experience within England during the pandemic thus far. METHODS: A rapid phenomenological approach using semistructured telephone interviews with ED clinical leads from across England was undertaken between 16 and 22 April 2020. Participants were recruited through purposeful sampling with sample size determined by data saturation. Departments from a wide range of geographic distribution and COVID-19 experience were included. Themes were identified and included if they met one of three criteria: demonstrating a consistency of experience between EDs, demonstrating a conflict of approach between emergency departments or encapsulating a unique solution to a common barrier. RESULTS: Seven clinical leads from type 1 EDs were interviewed. Thematic redundancy was achieved by the sixth interview, and one further interview was performed to confirm. Themes emerged in five categories: departmental reconfiguration, clinical pathways, governance and communication, workforce and personal protective equipment. CONCLUSION: This paper summarises learning and innovation from a cross-section of EDs during the first UK wave of the COVID-19 pandemic. Common themes centred around the importance of flexibility when reacting to an ever-changing clinical challenge, clear leadership and robust methods of communication. Additionally, experience in managing winter pressures helped inform operational decisions, and ED staff demonstrated incredible resilience in demanding working conditions. Subsequent surges of COVID-19 infections may occur within a more challenging context with no guarantee that there will be an associated reduction in A&E attendance or cessation of elective activity. Future operational planning must therefore take this into consideration.

Crisis clinical pathway for COVID-19.

The pandemic of COVID-19 has been particularly severe in the New York City area, which has had one of the highest concentrations of cases in the USA. In March 2020, the EDs of New York-Presbyterian Hospital, a 10-hospital health system in the region, began to experience a rapid surge in patients with COVID-19 symptoms. Emergency physicians were faced with a disease that they knew little about that quickly overwhelmed resources. A significant amount of attention has been placed on the problem of limited supply of ventilators and intensive care beds for critically ill patients in the setting of the ongoing global pandemic. Relatively less has been given to the issue that precedes it: the demand on resources posed by patients who are not yet critically ill but are unwell enough to seek care in the ED. We describe here how at one institution, a cross-campus ED physician working group produced a care pathway to guide clinicians and ensure the fair and effective allocation of resources in the setting of the developing public health crisis. This 'crisis clinical pathway' focused on using clinical evaluation for medical decision making and maximising benefit to patients throughout the system.

Close air support: enhancing emergency care in the COVID-19 pandemic.

The COVID-19 pandemic has taken the world by storm and overwhelmed healthcare institutions even in developed countries. In response, clinical staff and resources have been redeployed to the areas of greatest need, that is, intensive care units and emergency rooms (ER), to reinforce front-line manpower. We introduce the concept of close air support (CAS) to augment ER operations in an efficient, safe and scalable manner. Teams of five comprising two on-site junior ER physicians would be paired with two CAS doctors, who would be off-site but be in constant communication via teleconferencing to render real-time administrative support. They would be supervised by an ER attending. This reduces direct viral exposure to doctors, conserves precious personal protective equipment and allows ER physicians to focus on patient care. Medical students can also be involved in a safe and supervised manner. After 1 month, the average time to patient disposition was halved. General feedback was also positive. CAS improves efficiency and is safe, scalable and sustainable. It has also empowered a previously untapped group of junior clinicians to support front-line medical operations, while simultaneously protecting them from viral exposure. Institutions can consider adopting our novel approach, with modifications made according to their local context.

Strategic planning and response to COVID-19 in a London emergency department.

For many of us in emergency medicine, rising to the challenge of the COVID-19 crisis will be the single most exciting and challenging episode of our careers. Lessons have been learnt on how to make quick and effective changes without being hindered by the normal restraints of bureaucracy. Changes that would normally have taken months to years to implement have been successfully introduced over a period of several weeks. Although we have managed these changes largely by command and control, compassionate leadership has identified leaders within our team and paved the way for the future. This article covers the preparation and changes made in response to COVID-19 in a London teaching hospital.

Emergency department 'outbreak rostering' to meet challenges of COVID-19.

The COVID-19 outbreak has posed unique challenges to the emergency department rostering. Additional infection control, the possibility of quarantine of staff and minimising contact among staff have significant impact on the work of doctors in the emergency department. Infection of a single healthcare worker may require quarantine of close contacts at work. This may thus affect a potentially large number of staff. As such, we developed an Outbreak Response Roster. This Outbreak Response Roster had fixed teams of doctors working in rotation, each team that staff the emergency department in turn. Members within teams remained constant and were near equally balanced in terms of manpower and seniority of doctors. Each team worked fixed 12 hours shifts with as no overlapping of staff or staggering of shifts. Handovers between shifts were kept as brief as possible. All these were measures to limit interactions among healthcare workers. With the implementation of the roster, measures were also taken to bolster the psychological wellness of healthcare workers. With face-to-face contact limited, we also had to maintain clear, open channels for communication through technology and continue educating residents through innovative means.

Anatomy of resuscitative care unit: expanding the borders of traditional intensive care units.

Resuscitation lacks a place in the hospital to call its own. Specialised intensive care units, though excellent at providing longitudinal critical care, often lack the flexibility to adapt to fluctuating critical care needs. We offer the resuscitative care unit as a potential solution to ensure that patients receive appropriate care during the most critical hours of their illnesses. These units offer an infrastructure for resuscitation and can meet the changing needs of their institutions.

A simple clinical assessment is superior to systematic triage in prediction of mortality in the emergency department.

OBJECTIVE: To compare the Danish Emergency Process Triage (DEPT) with a quick clinical assessment (Eyeball triage) as predictors of short-term mortality in patients in the emergency department (ED). METHODS: The investigation was designed as a prospective cohort study conducted at North Zealand University Hospital. All patient visits to the ED from September 2013 to December 2013 except minor injuries were included. DEPT was performed by nurses. Eyeball triage was a quick non-systematic clinical assessment based on patient appearance performed by phlebotomists. Both triage methods categorised patients as green (not urgent), yellow, orange or red (most urgent). Primary analysis assessed the association between triage level and 30-day mortality for each triage method. Secondary analyses investigated the relation between triage level and 48-hour mortality as well as the agreement between DEPT and Eyeball triage. RESULTS: A total of 6383 patient visits were included. DEPT was performed for 6290 (98.5%) and Eyeball triage for 6382 (~100%) of the patient visits. Only patients with both triage assessments were included. The hazard ratio (HR) for 48-hour mortality for patients categorised as yellow was 0.9 (95% CI 0.4 to 1.9) for DEPT compared with 4.2 (95% CI 1.2 to 14.6) for Eyeball triage (green is reference). For orange the HR for DEPT was 2.2 (95% CI 1.1 to 4.4) and 17.1 (95% CI 5.1 to 57.1) for Eyeball triage. For red the HR was 30.9 (95% CI 12.3 to 77.4) for DEPT and 128.7 (95% CI 37.9 to 436.8) for Eyeball triage. For 30-day mortality the HR for patients categorised as yellow was 1.7 (95% CI 1.2 to 2.4) for DEPT and 2.4 (95% CI 1.6 to 3.5) for Eyeball triage. For orange the HR was 2.6 (95% CI 1.8 to 3.6) for DEPT and 7.6 (95% CI 5.1 to 11.2) for Eyeball triage, and for red the HR was 19.1 (95% CI 10.4 to 35.2) for DEPT and 27.1 (95% CI 16.9 to 43.5) for Eyeball triage. Agreement between the two systems was poor (kappa 0.05). CONCLUSION: Agreement between formalised triage and clinical assessment is poor. A simple clinical assessment by phlebotomists is superior to a formalised triage system to predict short-term mortality in ED patients.

Diagnostic accuracy of a rapid telemedicine encounter in the Emergency Department.

OBJECTIVES: Emergency Department crowding is an increasing problem, leading to treatment delays and higher risk of mortality. Our institution recently implemented a telemedicine physician intake ("tele-intake") process as a mitigating front-end strategy. Previous studies have focused on ED throughput metrics such as door to disposition; our work aimed to specifically assess the tele-intake model for clinical accuracy. METHODS: We retrospectively reviewed ED visits at a high acuity, tertiary care academic hospital before and after tele-intake implementation. We defined the primary outcome as the degree of additional laboratory, imaging, and medication orders placed by the subsequent ED provider. Our secondary outcomes were the cancellation rate of intake orders and the percentage of encounters where no additional second provider orders were necessary. RESULTS: For in-person and tele-intake physician encounters between September 2015 and February 2017, most labs and diagnostic radiology studies, and approximately half of CT, ultrasound, and pharmacy orders were initiated by the intake physician. We found no significant difference for our primary outcome (p = 0.2449). For both tele-intake and in-person encounters, <1% of orders were cancelled by the second provider. Additionally, 30.8% of in-person and 31.5% of telemedicine patient encounters required no additional orders to make a disposition decision. DISCUSSION: This novel analysis of an innovative patient care model suggests that the benefits of tele-intake as a replacement for in-person physician directed intake are not at the cost of over or under utilization of diagnostic testing or interventions.

Impact of Physician Navigators on productivity indicators in the ED.

OBJECTIVES: We created Physician Navigators in our ED to help improve emergency physician (EP) productivity. We aimed to quantify the effect of Physician Navigators on measures of EP productivity: patient seen per hour (Pt/hr), and turn-around time (TAT) to discharge. Secondary objectives included examining their impact on measures of ED throughput for non-resuscitative patients: ED length of stay (LOS), door-to-physician time and left-without-being-seen rates (LWBS). METHODS: In this retrospective study, 6845 clinical shifts worked by 20 EPs at a community ED in Newmarket, Canada from 1 January 2012 to 31 March 2015 were evaluated. Using a clustered design, we compared productivity measures between shifts with and without Physician Navigators, by physician. We used a linear mixed model to examine mean changes in Pt/hr and TAT to discharge for EPs who employed Physician Navigators. For secondary objectives, autoregressive modelling was performed to compare ED throughput metrics before and after the implementation of Physician Navigators for non-resuscitative patients. RESULTS: Patient volumes increased by 20 patients per day (p<0.001). Mean Pt/hr increased by 1.07 patients per hour (0.98 to 1.16, p<0.001). The mean TAT to discharge decreased by 10.6 min (-13.2 to -8.0, p<0.001). After implementation of the Physician Navigator programme, overall mean LOS for non-resuscitative patients decreased by 2.6 min (p=0.007), and mean door-to-physician time decreased by 7.4 min (p<0.001). LBWS rates decreased from 1.13% to 0.63% of daily patient volume (p<0.001). CONCLUSION: Despite an ED volume increase, the use of a Physician Navigator was associated with significant improvements in EP productivity, and significant reductions in ED throughput times.

Modelling attending physician productivity in the emergency department: a multicentre study.

OBJECTIVES: Emergency physician productivity, often defined as new patients evaluated per hour, is essential to planning clinical operations. Prior research in this area considered this a static quantity; however, our group's study of resident physicians demonstrated significant decreases in hourly productivity throughout shifts. We now examine attending physicians' productivity to determine if it is also dynamic. METHODS: This is a retrospective cohort study, conducted from 2014 to 2016 across three community hospitals in the north-eastern USA, with different schedules and coverage. Timestamps of all patient encounters were automatically logged by the sites' electronic health record. Generalised estimating equations were constructed to predict productivity in terms of new patients per shift hour. RESULTS: 207 169 patients were seen by 64 physicians over 2 years, comprising 9822 physician shifts. Physicians saw an average of 15.0 (SD 4.7), 20.9 (SD 6.4) and 13.2 (SD 3.8) patients per shift at the three sites, with 2.97 (SD 0.22), 2.95 (SD 0.24) and 2.17 (SD 0.09) in the first hour. Across all sites, physicians saw significantly fewer new patients after the first hour, with more gradual decreases subsequently. Additional patient arrivals were associated with greater productivity; however, this attenuates substantially late in the shift. The presence of other physicians was also associated with slightly decreased productivity. CONCLUSIONS: Physician productivity over a single shift follows a predictable pattern that decreases significantly on an hourly basis, even if there are new patients to be seen. Estimating productivity as a simple average substantially underestimates physicians' capacity early in a shift and overestimates it later. This pattern of productivity should be factored into hospitals' staffing plans, with shifts aligned to start with the greatest volumes of patient arrivals.