ED Wait Times Research Articles

Virtual triage from freestanding emergency departments: a propensity score-weighted analysis of short-term outcomes in emergency general surgery.

Freestanding emergency departments (FSEDs) have generated improved hospital metrics, including decreased ED wait times and increased patient selection. Patient outcomes and process safety have not been evaluated. This study investigates the safety of FSED virtual triage in the emergency general surgery (EGS) patient population. A retrospective review evaluated all adult EGS patients admitted to a community hospital between January 2016 and December 2021 who either presented at a FSED and received virtual evaluation from a surgical team (fEGS) or presented at the community hospital emergency department and received in-person evaluation from the same surgical group (cEGS). Patients' demographics, acute care utilization history, and clinical characteristics at the onset of the index visit were used to build a propensity score model and stabilized Inverse Probability of Treatment Weights (IPTW) were used to create a weighted sample. Multivariable regression models were then employed to the weighted sample to evaluate the treatment effect of virtual triage compared to in-person evaluation on short-term outcomes, including length of stay (LOS) and 30-day readmission and mortality. Variables which occurred during the index visit (such as surgery duration and type of surgery) were adjusted for in the multivariable analyses. Of 1962 patients, 631 (32.2%) were initially evaluated virtually (fEGS) and 1331 (67.8%) underwent an in-person evaluation (cEGS). Baseline characteristics demonstrated significant differences between the cohorts in gender, race, payer status, BMI, and CCI score. Baseline risks were well balanced in the IPTW-weighted sample (SD range 0.002-0.18). Multivariable analysis found no significant differences between the balanced cohorts in 30-day readmission, 30-day mortality, and LOS (p > 0.05 for all). Patients who undergo virtual triage have similar outcomes to those who undergo in-person triage for EGS diagnoses. Virtual triage at FSED for these EGS patients may be an efficient and safe means for initial evaluation.

Endogenous quality investments in the U.S. hospital market

High and increasing hospital prices could reflect market imperfections, including provider concentration. Yet high prices could also reflect increased willingness to pay by privately insured consumers for clinical and non-clinical quality. In this paper, we explore strategic quality choices where hospitals make quality investments to increase private revenue. We then measure the relationship between potential prices and numerous quality measures including patient satisfaction, hospital processes, risk-adjusted mortality, the revealed preferences of current Medicare patients, technology adoption, physician quality, and ED wait times. We show that across a range of measures quality is correlated with the profitability of the payer mix: hospitals with more potential privately insured patients are of higher quality.

Applying an Urgent Care Model to the Skilled Nursing Facility Setting

Reducing readmissions to hospitals and ED visits has been a priority among health systems and skilled nursing facilities. Approximately 20% of Medicare-funded hospitalizations result in discharge to a skilled nursing facility (SNF). Transitions to and from the hospital setting frequently occurs. In an analysis of SNF hospital readmissions 23% were found to be preventable. To reduce hospital readmissions from SNFs, we have deployed the Urgent Care model (UCM) within the SNF setting. Urgent care is typically defined as healthcare that is needed for a condition that requires prompt attention (within 24 h) but is not a life-threatening emergency. UCM are designed to manage unscheduled conditions, which often arise outside core office hours. The UCM has rapidly emerged over the last several decades out of demand to alleviate ED wait times and visits for patients. Several studies found that UCM have reduced visits to nearby EDs.

Facilitators and barriers to post-overdose service delivery in Rhode Island emergency departments: A qualitative evaluation

BackgroundEmergency departments (EDs) in the US have increasingly incorporated interventions that seek to reduce opioid-related morbidity and mortality. However, many of these interventions are underutilized. This study examined ED provider-identified barriers and facilitators to policy-prescribed service provision for patients treated for an opioid overdose in Rhode Island EDs, and opportunities to improve care delivery. MethodsSemi-structured qualitative interviews were conducted with 55 ED providers (management and clinical staff) across Rhode Island EDs from November 2019 to July 2020. Thematic analysis of interviews focused on gaps and best practices in post-overdose care delivery, including social and structural factors driving access to, and uptake of, services. ResultsParticipants highlighted how automatic service delivery (opt out vs. opt in) and the integration of peer-based services enhanced post-overdose service provision. However, social and structural factors (e.g. insurance barriers, limited outpatient treatment resources) and gaps in provider knowledge of medications for opioid use disorder created barriers to care. Addressing long ED wait times and establishing dedicated care teams for patients following an overdose were seen as critical to improving ED service delivery. ConclusionOur findings suggest that post-overdose service delivery within EDs is a useful approach for connecting patients to services, particularly when peer support specialists are involved. However, standardizing service delivery approaches and improving provider education of harm reduction services must be prioritized alongside state-level policy changes to improve access to care for ED patients.

Racial/ethnic disparities in emergency department wait times in the United States, 2013-2017.

Previous research shows that Black and Hispanic patients have longer ED wait times than White patients, but these data do not reflect recent changes such as the Affordable Care Act. In addition, previous research does not account for the non-normal distribution of wait times, wherein a sizable subgroup of patients seen promptly and those not seen promptly experience long wait times. We utilized National Hospital Ambulatory Medical Care Survey (NHAMCS) datasets (2013-2017) to examine mean ED wait time comparing visits by Black, Hispanic, and Asian patients to White patients. Using a two-part regression model, we adjusted for patient, hospital, and health system factors, and estimated differences, for each of five triage levels, in (a) likelihood of waiting at least 5 min and (b) difference in wait time among those not seen promptly. Our cohort included 38,800 White, 14,838 Black, 10,619 Hispanic, and 1257 Asian patient visits. Black (triage level 3) and Hispanic (triage levels 3 and 4) patients had longer mean wait times than White patients. Adjusted likelihood of not being seen promptly was lower among Blacks (triage levels 3, 4 and 5), Hispanics (triage level 5) and Asians (triage level 5) compared to Whites. Among those waiting at least 5 min, adjusted wait time was longer among Blacks in triage level 3 (5.2 min, 95% CI, 1.3 to 9.0) and level 4 (2.5 min, 95% CI, 0.2 to 4.9), Hispanics in triage level 4 (4.7 min, 95% CI, 1.7 to 7.7) and Asians in triage level 5 (16.3 min, 95% CI, 0.6 to 31.9) compared to Whites. Minority patients were less likely to wait to be seen, but waited longer if not seen promptly. These data exhibit that ED wait time disparities persist for African American and Hispanic patients and extend this observation to Asian patients.

P021: A novel way of hiding beds: manipulating wait time predictions to alter future patient flows into the ED

Introduction: Wait time predictions have become more common in emergency departments in Canada. These estimate the wait times a patient faces to see providers and they are usually provided in an accessible way such as through an online interface. One purpose of these trackers is to improve ED system efficiency. Patients can self-triage to alternative care such as their primary care physician, defer care until a later time or could move from oversubscribed to undersubscribed EDs. However, these mechanisms could also be abused. If providers can artificially influence the wait time this may provide a possible lever to change patients flows to an ED. I investigate whether there is evidence suggestive of manipulation of online wait time trackers at an ED system in Ontario. Methods: Inputs into the wait time prediction algorithm, like patient volumes are taken from the ED EMR. This is the most likely place where staff can manipulate the wait time tracker by retaining patients in the EMR system even after they are discharged. I examine two sets of data to assess whether the online tracker displays differences in patient volumes from “true” data. The first is scraped data of patient volumes from the wait times website. The second are the accurate patient volumes from administrative data which includes when a physician discharged patients from the ED. I compare values of the true patient volumes to the online values and plot distributions of these differences. I also employ measures of accuracy such as mean square error and root mean square error to provide a value of how accurate the online data is compared to the true data. I examine these by ED and over time. Results: There are differences between the number of patients that are posted online and those in the administrative data. The distributions of these differences are skewed towards positive values suggesting that the online data more often overcounts rather than undercounts patients. Measures of accuracy increase during times when EDs are congested but do not decrease when EDs become less congested. This inaccuracy persists for a period after EDs cease to be busy. Conclusion: ED wait time trackers have the potential to be manipulated. When staff have incentive to reduce patient volumes, online data becomes more inaccurate relative to true data. This suggests that wait time trackers may have unintended consequences and that the information that they provide may not be entirely accurate.

Endogenous Quality Investments in the U.S. Hospital Market

High and increasing hospital prices have led to calls for price regulation. If prices are high because of consolidation, regulating prices could enhance welfare. However, high prices could also reflect increased willingness to pay by privately insured consumers for clinical and non-clinical quality. If so, regulating prices could reduce quality. We present a model of strategic quality choice where hospitals make quality investments to increase private revenue. We confirm the model's predictions across numerous quality measures including patient satisfaction, hospital processes, risk adjusted mortality, the revealed preferences of current Medicare patients, technology adoption, physician quality, and ED wait times.

Effects of seasonality and daylight savings time on emergency department visits for mental health disorders

ObjectivesEmergency Department (ED) utilization accounts for a large portion of healthcare services in the US. Disturbance of circadian rhythms may affect mental and behavioral health (MBH) conditions, which could result in increased ED visits and subsequent hospitalizations, thus potentially inducing staffing shortages and increasing ED wait time. Predicting the burden of ED admissions helps to better plan care at the EDs and provides significant benefits. This study investigates if increased ED visits for MBH conditions are associated with seasonality and changes in daylight savings time. MethodsUsing ED encounter data from a large academic medical center, we have examined univariate and multivariate associations between ED visits for MBH conditions and the annual time periods during which MBH conditions are more elevated due to changes in the seasons. We hypothesize that ED visits for MBH conditions increase within the 2-week period following the daylight savings time changes. ResultsIncreased MBH ED visits were observed in certain seasons. This was especially true for non-bipolar depressive illness. We saw no significant changes in MBH visits as associated with changes in the daylight savings time. ConclusionsData do not provide conclusive evidence of a uniform seasonal increase in ED visits for MBH conditions. Variation in ED MBH visits may be due to secular trends, such as socioeconomic factors. Future research should explore contemporaneous associations between time-driven events and MBH ED visits. It will allow for greater understanding of challenges regarding psychiatric patients and opportunities for improvement.

Mortality outcomes and emergency department wait times – the paradox in the capacity limited system

Background: There is concern that undue ED wait times may result in adverse outcomes. Methods: We studied 30-day in-hospital mortality (2002-2017) for all medical admissions (106,586 episodes; 54,928 patients) focusing on clinical risk profile Results: Comparing 2002-09 vs. 2010-17, median ED waits > 6 hours (hr) increased 10h (95% CI: 8,13) to 15h (95% CI: 9,19). 30-day mortality declined 6.2% to 4.9%- (RRR- 20.8%/ NNT- 78). 30-day-mortality by ED wait: – < 4hr 6.6% (95% CI: 6.3%, 6.9%), 4-8hr 4.8% (95% CI: 4.6%, 5.0%), 8-12hr 4.3% (95% CI: 4.1%, 4.5%) or >=12hr 4.2% (95% CI: 3.9%, 4.5%). Conclusion: Admissions with shorter waits are overrepresented with high clinical acuity. Higher Risk Score patient with extended wait times had worse clinical outcomes.

MP27: Publishing emergency department wait times in the waiting room: implementation and evaluation of a co-designed patient centered solution

Introduction: Patients in our ED were dissatisfied with their waiting experience, which resulted in patient anxiety and complaints. In 8 months, we aimed to (1) improve patient satisfaction with the ED waiting experience from triage to physician initial assessment by a 15% improvement in patients who rate their experience very good/excellent on a Likert Scale, and (2) improve patient knowledge of ED wait time by a 50% increase in understanding on a Likert Scale. Methods: We co-designed a display with ED patients to notify those in the waiting room of their wait process and wait time. The intervention was selected after root cause diagnostics including: Fishbone exercise, Pareto Diagram, and Driver Diagram. The display was co-designed with ED patients and improved via PDSA cycles to establish information displayed and how to incorporate it into the waiting experience. After co-design, a low-fidelity display was piloted in the waiting room. Results: A family of measures were evaluated using patient/provider surveys and hospital data metrics. Outcome measures were (1) percentage of patients who rated their ED experience as very good/excellent on a Likert scale, and (2) patients who had a clear/very clear understanding of their wait time on a Likert scale. Process measures were the percentage of patients who (1) looked at the wait time display, and (2) felt they could communicate their wait time to others. Balancing measures were clerk/nurse satisfaction and self-reported interruptions of patients asking wait time. Outcomes were tracked using statistical process charts and run charts. Following display implementation, patient rating of their ED experience and patient understanding of wait time showed positive improvement. Clerks/nurses were also more satisfied with their jobs and self-reported interruptions decreased. Conclusion: A low-fidelity wait time display co-designed with patients improved patient satisfaction and understanding of ED wait times. We plan to develop an automated electronic display that resembles the low-fidelity display and evaluate the impact of the intervention on the established measures. This intervention has the potential to be sustainable, feasible for other EDs, and require minimal upkeep costs.

Abstract 118: Factors Associated With Hospital Length of Stay for Heart Failure

Objective: To identify which hospital indicators are related to hospital length of stay (LOS) for heart failure (HF) inpatients, and their associations. Methods: Data from Centers for Medicare & Medicaid Services from 2012 to 2015 and the Florida Agency for Healthcare Administration (AHCA) from April 2014 to June 2016 were analyzed. We explored numerous quality and hospital size indicators, which were collected for nonidentical patient populations, for their potential associations with hospital mean LOS of Florida HF inpatients. We applied machine learning methods to screen for metrics most important to predicting LOS. Next, we derived generalized linear models (GLM) using several key screened indicators and multiple imputations for missing data, while weighting hospitals’ metrics according to their numbers of HF inpatients. We considered a total of 175 general acute care hospitals in Florida, and a total of 178,987 inpatient visits to obtain the values of mean hospital LOS. Results: Relationships obtained from the GLM for factors with substantial associations with hospital HF LOS are: Larger values of hospital-wide total staff size, emergency department wait time, and MRSA rates were all significantly associated with larger hospital HF LOS values. Increases in total hospital staff size from 550 to 5924 (the lowest and highest deciles of the hospital staff size distribution), given a fixed value of the MRSA rate, were associated with an average increase in the hospital HF LOS from its mean value over all hospitals of 5.0 to 5.9 (± .1) days. Increases in the mean time spent in the ED for all admitted patients from 222 to 374 minutes (the lowest and highest deciles of the distribution among hospitals for this ED wait time) were associated with an average increase in the hospital HF LOS from its mean of 5.0 to 6.0 (± .1) days. The GLM coefficient estimates (p < .0001) show these associations with the average over Florida hospitals of HF LOS. Additional factors, such as the number of HF inpatients, were also positively associated (though less strongly) with HF LOS, and are highly correlated with the indicators above such as hospital staff size. Conclusion: For Florida inpatients, the mean hospital HF LOS was positively associated with hospital-wide metrics of total staff size, MRSA rate, and time spent in the ED. Since the data were observational and many factors were not controlled, potential causality can not be determined. Further study is needed to discover the mechanisms behind these associations, as well as any potential confounding factors. Disclaimers: Results here were derived from limited data sets of the Florida AHCA. However, Florida AHCA does not claim responsibility for any of the resulting analysis, interpretations or conclusions. The authors and their institutions do not claim that the associations found are causal, nor that any actions taken based on the results here will change outcome metrics.

A nurse-driven protocol for oral rehydration therapy in mildly dehydrated pediatric patients

A nurse-driven protocol for oral rehydration therapy in mildly dehydrated pediatric patients

Who breaches the four-hour emergency department wait time target? A retrospective analysis of 374,000 emergency department attendances between 2008 and 2013 at a type 1 emergency department in England

BackgroundThe four-hour target is a key hospital emergency department performance indicator in England and one that drives the physical and organisational design of the ED. Some studies have identified time of presentation as a key factor affecting waiting times. Few studies have investigated other determinants of breaching the four-hour target. Therefore, our objective was to describe patterns of emergency department breaches of the four-hour wait time target and identify patients at highest risk of breaching.MethodsThis was a retrospective cohort study of a large type 1 Emergency department at an NHS teaching hospital in Oxford, England. We analysed anonymised individual level patient data for 378,873 emergency department attendances, representing all attendances between April 2008 and April 2013. We examined patient characteristics and emergency department presentation circumstances associated with the highest likelihood of breaching the four-hour wait time target.ResultsWe used 374,459 complete cases for analysis. In total, 8.3% of all patients breached the four-hour wait time target. The main determinants of patients breaching the four-hour wait time target were hour of arrival to the ED, day of the week, patient age, ED referral source, and the types of investigations patients receive (p < 0.01 for all associations). Patients most likely to breach the four-hour target were older, presented at night, presented on Monday, received multiple types of investigation in the emergency department, and were not self-referred (p < 0.01 for all associations). Patients attending from October to February had a higher odds of breaching compared to those attending from March to September (OR 1.63, 95% CI 1.59 to 1.66).ConclusionsThere are a number of independent patient and circumstantial factors associated with the probability of breaching the four-hour ED wait time target including patient age, ED referral source, the types of investigations patients receive, as well as the hour, day, and month of arrival to the ED. Efforts to reduce the number of breaches could explore late-evening/overnight staffing, access to diagnostic tests, rapid discharge facilities, and early assessment and input on diagnostic and management strategies from a senior practitioner.

Commentaries on health services research

Commentaries on health services research

LO17: A comparative evaluation of ED crowding metrics and associations with patient mortality

Introduction: Over 700 different input, throughput and output metrics have been used to quantify ED crowding. Of these, only ED length-of-stay (ED LOS) has been shown to be associated with mortality. No comparative evaluation of ED crowding metrics has been performed to determine which ones have the strongest association with patient mortality. The objective of this study was to compare the strength of association of common ED input, throughput and output metrics to patient mortality. Methods: Administrative data from five years of ED visits (2011-2014) at three urban EDs were linked to develop a database of over 900,000 ED visits with patient demographics, electronic time stamps for care processes, dispositions and outcomes. The data were randomly divided into three partitions of equal size. Here we report the findings from one partition of 253,938 ED visits. The remaining two data partitions will be used to validate these findings. Commonly-used crowding metrics were quantified and aggregated by day or by shift (0800-1600, 1600-2400, 2400-0800), and the shift-specific metrics assigned to each patient. The primary outcome was 7-day all-cause mortality. Multilevel logistic regression models were developed for 7-day mortality, with selected ED crowding metrics and a common set of confounders as predictors. The strength of association between the crowding metrics and mortality was compared using Akaike’s Information Criterion (AIC) and the Bayesian Information Criterion (BIC): ED crowding metrics with lower AIC and BIC have stronger associations with 7-day mortality. Results: Of 909,000 ED encounters, 124,679 (16.5%) arrived by EMS, 149,233(19.7%) were admitted, and 3,808 patients (0.5%) died within 7 days of ED arrival. Of input metrics, the model with ED wait-time was better (i.e. had a smaller AIC and BIC) than models for daily census, ED occupancy or LWBS proportion for predicting 7-day mortality. Of throughput metrics, the model with mean ED LOS was better than the model for mean MD care time. Of output metrics, the model with daily inpatient hospital occupancy was better than the model with mean boarding time. Conclusion: Based on one data partition, regression models based on the average wait-time, ED LOS and inpatient occupancy best predicted 7-day mortality. These results will be validated in the two other data partitions to confirm the best-performing ED input, throughput and output metrics.

Nurse-Initiated Protocols Can Reduce ED Wait Times.

Nurse-Initiated Protocols Can Reduce ED Wait Times.

At Your Defense

At Your Defense

IMPACT OF CENTRALIZED REMOTE MONITORING ON THE EFFICIENCY OF CARDIAC DEVICE PATIENT CARE

Over 250,000 Canadians are living with Cardiac Implantable Electrical Devices (CIEDs). These devices improve survival & reduce morbidity, but their follow-up is resource intensive; both in terms of the healthcare system & recipients (travel, parking & time). Remote Monitoring (RM), CIED follow-up in a patient’s home or a centralized area, is an important adjunct to conventional follow-up. Home-based RM has been shown to reduce cost & enhance quality of life without compromising safety. However, centralized RM is less well studied. Centralized RM uses a single device to collect the data from multiple CIEDs centrally (e.g., an emergency department or ED) for the Clinician to review and interpret, similar to a home-based RM encounter. To assess the feasibility, acceptance and impact of CareLink Express[TM], a centralized RM tool, in a large urban region. CareLink Express was implemented at 4 high volume EDs in a large metropolitan area. ED staff identified CIED patients that required evaluation, an ECG technician performed CIED interrogation via CareLink Express & the CIED Clinician on call was notified. The Clinician then reviewed the interrogation from home & reported the results to ED staff and the on call electrophysiologist, as necessary. A total 243 evening and nightshift CareLink Express assessments occurred over 12 months. Most assessments did not require direct assessment by the CIED clinician or electrophysiologist. CareLink Express assessments eliminated an estimated 486 hours of travel and 1,458 hours of Clinician overtime. This resulted in a savings of $54,171. Clinician assessment of CareLink Express was universally positive, with reduced travel time cited as its’ main advantage. ECG technician satisfaction was also high. ED length of stay was estimated to be reduced by an average of nearly 4 hours per CareLink Express assessment (total reduction of 785 hours). This reduced ED wait times and resulted in an estimated further cost savings of $51,679. Reliable, comprehensive estimates of costs are complex since our implementation of this tool involved multiple departments, clinics & hospitals. The sources required for costs are being collated and will be available at the time of presentation. Centralized remote monitoring via CareLink Express was easy to implement, perceived as valuable and user friendly & allowed for efficient collection of CIED data. This process reduced staffing costs & travel plus enhanced the care of patients in the ED, resulting in improved ED bed utilization and additional savings.

Utilizing TQM and CQI prinicples in establishing a Type 1 observation unit in a comprehensive cancer center.

96 Background: CMS defines Observation Services as outpatient care ordered by a physician and provided in a hospital bed to determine the need for inpatient admission or discharge. Recent CMS rules on Observation stay and 2-midnight rule has created confusion amongst Medicare beneficiaries, providers and hospitals. Two-thirds of US hospitals deliver observation care without using an observation unit. Review of our hospital observation care and a survey of Cancer Centers within the Alliance of Dedicated Cancer Centers revealed that observation care was provided in a bed anywhere in the hospital with unstructured care provided at the discretion of the treating physician. This leads to less favorable outcomes compared to protocol driven and closed observation units (Type 1). Methods: Plan- Institutional stakeholders consisting of administrators, hospitalists, ED physicians, case managers and nurse managers developed an observation services guideline policy and project team. Do- pilot initiated with ED admitting Observation patients to dedicated hospitalist providers and unit. Order sets were developed, education and staff training was provided and target start date set. Baseline and performance measures were developed. Study- impact on ED flow, admissions, ED length of stay, provider satisfaction, diagnoses, patient case-mix, and financial costs. Act– Implemented changes and lessons learned from pilot Results: Before implementation of pilot less than 10 patient per month were placed in observation status (Sept 2011- Aug 2012). After pilot average of 30 patient per month were initially placed in Observation (Sept 2012-April 2014). Approximately 73% of patients placed on observation are discharged and not admitted. Observation counts for approximately 25,000 dollars in professional charges per month. ED LOS decreased to less than 6-hours. Patient satisfaction and ED provider satisfaction scores have improved. Conclusions: Majority of Observation patients are not admitted thus avoiding the higher cost associated with inpatient stays with resulting cost saving. ED wait times have improved leading to less over-crowding with ED physicians having time to see more patients.

Using Google Flu Trends data in forecasting influenza-like–illness related ED visits in Omaha, Nebraska

IntroductionEmergency department (ED) visits increase during the influenza seasons. It is essential to identify statistically significant correlates in order to develop an accurate forecasting model for ED visits. Forecasting influenza-like–illness (ILI)-related ED visits can significantly help in developing robust resource management strategies at the EDs. MethodsWe first performed correlation analyses to understand temporal correlations between several predictors of ILI-related ED visits. We used the data available for Douglas County, the biggest county in Nebraska, for Omaha, the biggest city in the state, and for a major hospital in Omaha. The data set included total and positive influenza test results from the hospital (ie, Antigen rapid (Ag) and Respiratory Syncytial Virus Infection (RSV) tests); an Internet-based influenza surveillance system data, that is, Google Flu Trends, for both Nebraska and Omaha; total ED visits in Douglas County attributable to ILI; and ILI surveillance network data for Douglas County and Nebraska as the predictors and data for the hospital's ILI-related ED visits as the dependent variable. We used Seasonal Autoregressive Integrated Moving Average and Holt Winters methods with3 linear regression models to forecast ILI-related ED visits at the hospital and evaluated model performances by comparing the root means square errors (RMSEs). ResultsBecause of strong positive correlations with ILI-related ED visits between 2008 and 2012, we validated the use of Google Flu Trends data as a predictor in an ED influenza surveillance tool. Of the 5 forecasting models we have tested, linear regression models performed significantly better when Google Flu Trends data were included as a predictor. Regression models including Google Flu Trends data as a predictor variable have lower RMSE, and the lowest is achieved when all other variables are also included in the model in our forecasting experiments for the first 5 weeks of 2013 (with RMSE = 57.61). ConclusionsGoogle Flu Trends data statistically improve the performance of predicting ILI-related ED visits in Douglas County, and this result can be generalized to other communities. Timely and accurate estimates of ED volume during the influenza season, as well as during pandemic outbreaks, can help hospitals plan their ED resources accordingly and lower their costs by optimizing supplies and staffing and can improve service quality by decreasing ED wait times and overcrowding.