Alternatives to Opioids in the Emergency Department: The Henry Ford Health System Experience

Study Objectives: The aim of our Alternatives to Opiates (ALTO) program was to decrease opioid administration, measured in morphine milliequivalents (MME) per patient encounter, in all Henry Ford Health System (HFHS) emergency departments (EDs) by 15% in the 1-year period after program implementation as compared to the 3-year MME per patient encounter baseline prior to implementation. To our knowledge, this is the largest evaluation of an ALTO protocol completed. Methods: A multi-disciplinary group of ED providers, pharmacists, and nursing developed an ALTO program for HFHS using existing best practices and publications for implementation at all 9 HFHS EDs. This included 5 hospitals and 4 free-standing EDs, which see approximately 450,000 patient visits annually in total. Henry Ford Hospital (HFH) ED, a 100,000 annual patient visit quaternary care center, was amongst the 9 EDs included. After finalization of a protocol, we implemented an ALTO “Quicklist,” an organized section of non- opiate pain medications for indicated conditions, into our electronic medical record (EMR, Figure 1). Prior to implementation of our ALTO program in November 2019, we provided education surrounding the new tools and protocols to EM providers and nursing staff in different forums including staff meetings, grand rounds, nursing huddles and also email. Feedback was provided to all departments via system meetings and email communication at 4-month and 8-month intervals post-program kick off to monitor progress. A list of all opioids on HFHS pharmaceutical formularies was reviewed and a standard conversion factor to MME was applied to calculate a total number of MME administered per patient encounter (Figure 2). Only opiate medications administered in the ED were included. MME per patient encounter were compared in two groups: pre-intervention (November 2016 through October 2019;n=1, 317, 466) to the 1-year period post implementation (December 2019 through November 2020;n=366, 404). MME per patient encounter were compared for these time periods across HFHS EDs and for each department and the percentage change was calculated. Results: Opiate administration decreased across the entire HFHS from 2.76 MME per patient encounter to 2.62 MME per patient encounter, a 5.1% decrease overall. Only one of 9 EDs did not see any decrease. The other 8 EDs ranged from 0.5% to 29.4% decrease. Sub-analysis showed opiate administration at the HFH ED decreased from 4.60 MME per patient encounter to 4.28 MME per patient encounter, a 6.8% decrease. Conclusions: The HFHS ALTO program decreased opiate administration across the entire system, however, the goal of a 15% reduction in opiate administration was not met. The COVID-19 pandemic likely confounded these results with an increase in acuity and length of boarding in the ED juxtaposed against a decrease in patient volume seen across the country. Additionally, opiate administration was not stratified among discharged, admitted and intubated patients. Continued ALTO program education is ongoing as is further study with more nuanced data analysis, including specific ALTO medication utilization and provider level data. Other departments in the system, such as observation and surgical specialties, have expressed interest in developing similar protocols and QuickLists. [Formula presented] [Formula presented]

Forecasting Daily Patient Arrivals during COVID-19 in Emergency Departments: A Deep Learning Approach

Study Objectives: As the fourth wave of coronavirus disease 2019 (COVID-19) surges in Michigan, most health care systems are experiencing an increased hospitalization rate of infected COVID-19 patients. Understanding the arrival rates of patients to the emergency department (ED) is fundamental in managing the limited health care resources. Our objective is to develop an accurate forecasting model based on ED patients’ arrival and COVID-19 status to help manage and facilitate a data-driven resource planning. Methods: A cohort study of patients with clinical suspicion of COVID-19 evaluated at 2 EDs within an integrated health system that cares for a racially diverse population. We included patient arrivals, COVID-19 status, and demographic information between the dates of January 1, 2020 and March 16, 2021. We developed deep learning models (Long Short-Term Memory (LSTM)) to forecast patient arrivals in two geographically diverse EDs (denoted as ED1 and ED2). We used data from January to December 2020 for model training and data from January 2021 to March 2021 for model validation. The models are evaluated based on the root mean squared error (RMSE), the square root of the average of the squared error between predicted and observed values, and the mean absolute error (MAE), which provides the mean absolute difference between the predicted and the observed ED patient arrival rates per day. Results: In ED1, there were 56, 61 total patient arrivals (1, 433 infected COVID-19 patients) with a mean age of 38.0 ± 21.2 years. A majority were female (33, 457, 59.1%) and 29, 040 (51.3%) were Black. The average patient arrival per day was 125.1 (SD 35.0) for those without COVID-19, and 3.3 (SD 3.6) for COVID-1 confirmed patients. In ED2, there were 74, 176 total patient arrivals (1, 546 infected COVID-19 patients) with a mean age of 45.0 ± 23.0 years. A majority were female (39, 521, 53.3%) and 10, 636 (14.3%) were Black. The average patient arrival per day was 164.7 (SD 33.2) for those without COVID-19, and 3.5 (SD 5.0) for COVID 19 confirmed patients. Figure 1 shows the observed and predicted patients’ arrival for the two EDs for regular and confirmed COVID-19 patients. The LSTM models show accurate prediction one week in advance of daily patient arrivals for ED1 and ED2 with RMSE scores of 17 and 20 patients, respectively. The MAE values imply that, on average, the forecast’s error from the true daily patient arrival rate is 13.9 and 16.0 for ED1 and ED2, respectively. For COVID-19 patient arrivals to ED1 and ED2, the RMSE score is 3 patients each, while th MAE values are 2.2 and 2.4, respectively. Conclusion: This study demonstrates that an average RMSE prediction score of 18.5 and 3 patient arrivals per day for regular and COVID-19 confirmed patients is possible across EDs using LSTM one week prior to forecasting. Future validation and implementation of such forecasting models could impact effective planning and allocation of limited ED and hospital resources. [Formula presented]

280 Boucebacks during the Covid-19 Pandemic

Study Objectives: During the pandemic, emergency clinicians balanced the growing crisis of limited hospital bed availability with the risks of sending sick patients home. We sought to measure the rates of return visits during the pandemic and assess patient characteristics associated with higher rates of return. Methods: Cohort study of patients evaluated at 9 EDs within an integrated health system between March 13 and May 20, 2020 with clinical suspicion for Covid-19. We excluded patients who neither had testing for SARS-CoV-2 nor were designated with isolation precautions for Covid-19. We identified and collected data through a central dashboard that was established within the EHR. We defined confirmed Covid-19 cases as those with a positive PCR for SARS-CoV-2 infection. All patients had a minimum follow-up period of 14 days. The primary outcome was a return visit over the first 14 days. The analysis consisted of descriptive statistics and a multivariable proportional hazards model that was limited to patients discharged home on their index visit to assess the association between confirmed Covid-19 and bounceback. Results: There were 13,367 ED patients with clinical suspicion of Covid-19, of whom 7289 (54.5%) were female, 5225 (39.1%) black, non-Hispanic, and the mean age was 55.7 ±19.9 years. There were 12859 (96.2%) patients tested with PCR for SARS-CoV-2, 508 (3.8%) isolated for Covid-19 but never tested, and 3760 (28.1%) with confirmed Covid-19. The number of patients hospitalized was 7724 (57.8%). Return visits among those that were not hospitalized occurred 436 (7.7%) times within 14 days from the initial encounter and 546 (9.7%) times within 30 days. The median time to a return visit was 7 [IQR 3, 17] days. Of patients with a return visit in 14-days, 207 (46.1%) were hospitalized on their second visit. Patients who were discharged home that had confirmed Covid-19 had a return rate of 20.0% vs. 3.7% among patients without confirmed Covid-19 (see Figure 1). In multivariable analysis, factors not associated with the primary outcome were race, pulse oximetry, and sex. Factors significantly associated with 14-day returns were age >60 years (HR 1.34, 95% CI 1.03 - 1.67), each 1-point increase in the Charlson comorbidity index (HR 1.13, 95% CI 1.03 - 1.17), and confirmed Covid-19 (HR 5.25, 95% CI 4.29 - 6.42). Conclusions: Admission rates were high in patients with suspected Covid-19, and return rates over 14 days were 7.7%. Patients with confirmed Covid-19 had a 5-fold greater hazard of a 14-day return compared to those without confirmed Covid-19. [Formula presented]