Patient sex, racial and ethnic disparities in emergency department triage: A multi-site retrospective study.

OBJECTIVES: There is limited evidence on sex, racial, and ethnic disparities in Emergency Department (ED) triage across diverse settings. We evaluated differences in the assignment of Emergency Severity Index (ESI) by patient sex and race/ethnicity, accounting for age, clinical factors, and ED operating conditions. METHODS: We conducted a multi-site retrospective study of adult patients presenting to high-volume EDs from January 2019-February 2020. Patient-level data were obtained and analyzed from three EDs (academic, metropolitan community, and rural community) affiliated with a large health system in the Southeastern United States. For the study outcome, ESI levels were grouped into three categories: 1-2 (highest acuity), 3, and 4-5 (lowest acuity). Multinomial logistic regression was used to compare ESI categories by patient race/ethnicity and sex jointly (referent = White males), adjusted for patient age, insurance status, ED arrival mode, chief complaint category, comorbidity score, time of day, day of week, and average ED wait time. RESULTS: We identified 186,840 eligible ED visits with 56,417 from the academic ED, 69,698 from the metropolitan community ED, and 60,725 from the rural community ED. Patient cohorts between EDs varied by patient age, race/ethnicity, and insurance status. The majority of patients were assigned ESI 3 in the academic and metropolitan community EDs (61% and 62%, respectively) whereas 47% were assigned ESI 3 in the rural community ED. In adjusted analyses, White females were less likely to be assigned ESI 1-2 compared to White males although both groups were roughly comparable in the assignment of ESI 4-5. Non-White and Hispanic females were generally least likely to be assigned ESI 1-2 in all EDs. Interactions between ED wait time and race/ethnicity-sex were not statistically significant. CONCLUSIONS: This retrospective study of adult ED patients revealed sex and race/ethnicity-based differences in ESI assignment, after accounting for age, clinical factors, and ED operating conditions. These disparities persisted across three different large EDs, highlighting the need for ongoing research to address inequities in ED triage decision-making and associated patient-centered outcomes.

Disparities in emergency department prioritization and rooming of patients with similar triage acuity score.

BACKGROUND: We identify patient demographic and emergency department (ED) characteristics associated with rooming prioritization decisions among ED patients who are assigned the same triage acuity score. METHODS: We performed a retrospective analysis of adult ED patients with similar triage acuity, as defined as an Emergency Severity Index (ESI) of 3, at a large academic medical center, during 2019. Violations of a first-come-first-served (FCFS) policy for rooming are identified and used to create weighted multiple logistic regression models and 1:M matched case-control conditional logistic regression models to determine how rooming prioritization is affected by individual patient age, sex, race, and ethnicity after adjusting for patient clinical and time-varying ED operational characteristics. RESULTS: A total of 15,781 ED encounters were analyzed, with 1612 (10.2%) ED encounters having a rooming prioritization in violation of a FCFS policy. Patient age and race were found to be significantly associated with being prioritized in violation of FCFS in both logistic regression models. The 1:M matched model showed a statistically significant relationship between violation of rooming prioritization with increasing age in years (adjusted odds ratio [aOR] 1.009, 95% confidence interval [CI] 1.005-1.013) and among African American patients compared to Caucasians (aOR 0.636, 95% CI 0.545-0.743). CONCLUSIONS: Among ED patients with a similar triage acuity (ESI 3), we identified patient age and patient race as characteristics that were associated with deviation from a FCFS prioritization in ED rooming decisions. These findings suggest that there may be patient demographic disparities in ED rooming decisions after adjusting for clinical and ED operational characteristics.

Forecasting emergency department hourly occupancy using time series analysis.

STUDY OBJECTIVE: To develop a novel predictive model for emergency department (ED) hourly occupancy using readily available data at time of prediction with a time series analysis methodology. METHODS: We performed a retrospective analysis of all ED visits from a large academic center during calendar year 2012 to predict ED hourly occupancy. Due to the time-of-day and day-of-week effects, a seasonal autoregressive integrated moving average with external regressor (SARIMAX) model was selected. For each hour of a day, a SARIMAX model was built to predict ED occupancy up to 4-h ahead. We compared the resulting model forecast accuracy and prediction intervals with previously studied time series forecasting methods. RESULTS: The study population included 65,132 ED visits at a large academic medical center during the year 2012. All adult ED visits during the first 265 days were used as a training dataset, while the remaining ED visits comprised the testing dataset. A SARIMAX model performed best with external regressors of current ED occupancy, average department-wide ESI, and ED boarding total at predicting up to 4-h-ahead ED occupancy (Mean Square Error (MSE) of 16.20, and 64.47 for 1-hr- and 4-h- ahead occupancy, respectively). Our 24-SARIMAX model outperformed other popular time series forecasting techniques, including a 60% improvement in MSE over the commonly used rolling average method, while maintaining similar prediction intervals. CONCLUSION: Accounting for current ED occupancy, average department-wide ESI, and boarding total, a 24-SARIMAX model was able to provide up to 4 h ahead predictions of ED occupancy with improved performance characteristics compared to other forecasting methods, including the rolling average. The prediction intervals generated by this method used data readily available in most EDs and suggest a promising new technique to forecast ED occupancy in real time.

The effects of emergency department crowding on triage and hospital admission decisions.

BACKGROUND: Emergency department (ED) crowding is a recognized issue and it has been suggested that it can affect clinician decision-making. OBJECTIVES: Our objective was to determine whether ED census was associated with changes in triage or disposition decisions made by ED nurses and physicians. METHODS: We performed a retrospective study using one year of data obtained from a US academic center ED (65,065 patient encounters after cleaning). Using a cumulative logit model, we investigated the association between a patient's acuity group (low, medium, and high) and ED census at triage time. We also used multivariate logistic regression to investigate the association between the disposition decision for a patient (admit or discharge) and the ED census at the disposition decision time. In both studies, control variables included census, age, gender, race, place of treatment, chief complaint, and certain interaction terms. RESULTS: We found statistically significant correlation between ED census and triage/disposition decisions. For each additional patient in the ED, the odds of being assigned a high acuity versus medium or low acuity at triage is 1.011 times higher (95% confidence interval [CI] for Odds Ratio [OR] = [1.009,1.012]), and the odds of being assigned medium or high acuity versus low acuity at triage is 1.009 times higher (95% CI for OR = [1.008,1.010]). Similarly, the odds of being admitted versus discharged increases by 1.007 times (95% CI for OR = [1.006,1.008]) per additional patient in the ED at the time of disposition decision. CONCLUSION: Increased ED occupancy was found to be associated with more patients being classified as higher acuity as well as higher hospital admission rates. As an example, for a commonly observed patient category, our model predicts that as the ED occupancy increases from 25 to 75 patients, the probability of a patient being triaged as high acuity increases by about 50% and the probability of a patient being categorized as admit increases by around 25%.

Comparison of emergency department crowding scores: a discrete-event simulation approach.

According to American College of Emergency Physicians, emergency department (ED) crowding occurs when the identified need for emergency services exceeds available resources for patient care in the ED, hospital, or both. ED crowding is a widely reported problem and several crowding scores are proposed to quantify crowding using hospital and patient data as inputs for assisting healthcare professionals in anticipating imminent crowding problems. Using data from a large academic hospital in North Carolina, we evaluate three crowding scores, namely, EDWIN, NEDOCS, and READI by assessing strengths and weaknesses of each score, particularly their predictive power. We perform these evaluations by first building a discrete-event simulation model of the ED, validating the results of the simulation model against observations at the ED under consideration, and utilizing the model results to investigate each of the three ED crowding scores under normal operating conditions and under two simulated outbreak scenarios in the ED. We conclude that, for this hospital, both EDWIN and NEDOCS prove to be helpful measures of current ED crowdedness, and both scores demonstrate the ability to anticipate impending crowdedness. Utilizing both EDWIN and NEDOCS scores in combination with the threshold values proposed in this work could provide a real-time alert for clinicians to anticipate impending crowding, which could lead to better preparation and eventually better patient care outcomes.

ReSTART: A Novel Framework for Resource-Based Triage in Mass-Casualty Events.

OBJECTIVE: Current guidelines for mass-casualty triage do not explicitly use information about resource availability. Even though this limitation has been widely recognized, how it should be addressed remains largely unexplored. The authors present a novel framework developed using operations research methods to account for resource limitations when determining priorities for transportation of critically injured patients. To illustrate how this framework can be used, they also develop two specific example methods, named ReSTART and Simple-ReSTART, both of which extend the widely adopted triage protocol Simple Triage and Rapid Treatment (START) by using a simple calculation to determine priorities based on the relative scarcity of transportation resources. METHODS: The framework is supported by three techniques from operations research: mathematical analysis, optimization, and discrete-event simulation. The authors? algorithms were developed using mathematical analysis and optimization and then extensively tested using 9,000 discrete-event simulations on three distributions of patient severity (representing low, random, and high acuity). For each incident, the expected number of survivors was calculated under START, ReSTART, and Simple-ReSTART. A web-based decision support tool was constructed to help providers make prioritization decisions in the aftermath of mass-casualty incidents based on ReSTART. RESULTS: In simulations, ReSTART resulted in significantly lower mortality than START regardless of which severity distribution was used (paired t test, p<.01). Mean decrease in critical mortality, the percentage of immediate and delayed patients who die, was 8.5% for low-acuity distribution (range ?2.2% to 21.1%), 9.3% for random distribution (range ?0.2% to 21.2%), and 9.1% for high-acuity distribution (range ?0.7% to 21.1%). Although the critical mortality improvement due to ReSTART was different for each of the three severity distributions, the variation was less than 1 percentage point, indicating that the ReSTART policy is relatively robust to different severity distributions. CONCLUSIONS: Taking resource limitations into account in mass-casualty situations, triage has the potential to increase the expected number of survivors. Further validation is required before field implementation; however, the framework proposed in here can serve as the foundation for future work in this area.

A spoonful of math helps the medicine go down: an illustration of how healthcare can benefit from mathematical modeling and analysis.

OBJECTIVES: A recent joint report from the Institute of Medicine and the National Academy of Engineering, highlights the benefits of--indeed, the need for--mathematical analysis of healthcare delivery. Tools for such analysis have been developed over decades by researchers in Operations Research (OR). An OR perspective typically frames a complex problem in terms of its essential mathematical structure. This article illustrates the use and value of the tools of operations research in healthcare. It reviews one OR tool, queueing theory, and provides an illustration involving a hypothetical drug treatment facility. METHOD: Queueing Theory (QT) is the study of waiting lines. The theory is useful in that it provides solutions to problems of waiting and its relationship to key characteristics of healthcare systems. More generally, it illustrates the strengths of modeling in healthcare and service delivery.Queueing theory offers insights that initially may be hidden. For example, a queueing model allows one to incorporate randomness, which is inherent in the actual system, into the mathematical analysis. As a result of this randomness, these systems often perform much worse than one might have guessed based on deterministic conditions. Poor performance is reflected in longer lines, longer waits, and lower levels of server utilization.As an illustration, we specify a queueing model of a representative drug treatment facility. The analysis of this model provides mathematical expressions for some of the key performance measures, such as average waiting time for admission. RESULTS: We calculate average occupancy in the facility and its relationship to system characteristics. For example, when the facility has 28 beds, the average wait for admission is 4 days. We also explore the relationship between arrival rate at the facility, the capacity of the facility, and waiting times. CONCLUSIONS: One key aspect of the healthcare system is its complexity, and policy makers want to design and reform the system in a way that affects competing goals. OR methodologies, particularly queueing theory, can be very useful in gaining deeper understanding of this complexity and exploring the potential effects of proposed changes on the system without making any actual changes.