A Global Survey of Emergency Department Responses to the COVID-19 Pandemic

Introduction Emergency departments (ED) globally are addressing the coronavirus disease 2019 (COVID-19) pandemic with varying degrees of success. We leveraged the 17-country, Emergency Medicine Education & Research by Global Experts (EMERGE) network and non-EMERGE ED contacts to understand ED emergency preparedness and practices globally when combating the COVID-19 pandemic. Methods We electronically surveyed EMERGE and non-EMERGE EDs from April 3–June 1, 2020 on ED capacity, pandemic preparedness plans, triage methods, staffing, supplies, and communication practices. The survey was available in English, Mandarin Chinese, and Spanish to optimize participation. We analyzed survey responses using descriptive statistics. Results 74/129 (57%) EDs from 28 countries in all six World Health Organization global regions responded. Most EDs were in Asia (49%), followed by North America (28%), and Europe (14%). Nearly all EDs (97%) developed and implemented protocols for screening, testing, and treating patients with suspected COVID-19 infections. Sixty percent responded that provider staffing/back-up plans were ineffective. Many sites (47/74, 64%) reported staff missing work due to possible illness with the highest provider proportion of COVID-19 exposures and infections among nurses. Conclusion Despite having disaster plans in place, ED pandemic preparedness and response continue to be a challenge. Global emergency research networks are vital for generating and disseminating large-scale event data, which is particularly important during a pandemic.


INTRODUCTION
Emergency departments (ED) globally are on the front lines in addressing the COVID-19 pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Most EDs have disaster preparedness plans in place for health system responses to large-scale disasters, including infectious Global Survey of ED Responses to COVID- 19 Mahajan et al. to poor health outcomes. Critical evaluation of the global response to the 2009 H1N1 influenza pandemic identified substantial variability and poorly conceived or even absent preparedness plans in many emergency care systems. 1 This missed opportunity to implement successful disaster response plans prior to the following major infectious outbreak highlights the need to study global ED responses and healthcare system preparedness on a continuous basis. 2,3 With the ongoing COVID-19 pandemic, knowledge regarding presentation, prognosis, and response to therapies continues to evolve. It is imperative for data, lessons learned, and successful approaches used by EDs with a high pandemic burden to be made rapidly and reliably available to those in earlier stages. Emergency medicine networks are valuable for collection of data supporting research, administrative, and educational goals and can potentially be leveraged to collate and disseminate experiences from disasters. 4 Emergency Medicine Education & Research by Global Experts (EMERGE) is a newly developed network of 26 EDs across 17 countries and six continents whose goal is to improve the care of acutely ill and injured patients by garnering the collective experiences of its member EDs. 5,6 In this study we sought to leverage the EMERGE network and establish collaborations with non-EMERGE EDs to determine global ED preparedness for COVID-19 and, specifically, to identify successful processes and protocols that may be adopted and/or adapted by other EDs to improve patient outcomes.

Study Design and Survey Development
We performed a cross-sectional study of ED practices during the pandemic using a survey sent to all participating sites. We reported our results using elements from the Strengthening the Reporting of Observational Studies in Epidemiology; Survey Reporting Guidelines; and the Checklist for Reporting Results of Internet E-Surveys guidelines. [7][8][9][10] We used the online survey platform Qualtrics (Qualtrics LLC, Provo, Utah) to characterize ED COVID-19 responses between April 3-June 1, 2020 along the following domains: pandemic preparedness plans and training; physical space; triage methods unique to the pandemic; staffing; supplies; and communication practices. The survey (Appendix A) was piloted and assessed for response process validity. Eleven emergency physicians at the University of Michigan evaluated the survey for construct and face-validity, and iteratively refined the questions. Survey questions were mainly closed-ended, with additional open-ended questions allowing participants to express opinions or to provide clarification on responses. The survey was translated using formal translation services into Mandarin Chinese and Spanish and reassessed for face validity in the translated language by EMERGE Executive Committee members who spoke those languages. The study was determined not to require regulation as human subjects research by the University of Michigan Institutional Review Board (HUM00178847).

Participants and Survey Distribution
Survey participation was voluntary, and respondents did not receive financial remuneration. The survey was distributed to EMERGE member institutions via email with unique links created for each participant. Attempts were made to involve non-EMERGE institutions by forwarding a one-page infographic (Appendix B) to contacts of EMERGE members and associates. Simultaneously, we contacted other international emergency associations for participation, including the Pan-Asian Resuscitation Outcomes Study, the World Academic Congress of Academic Medicine, the Michigan Emergency Department Improvement Collaborative, and the International Federation for Emergency Medicine. [11][12][13][14] To encourage survey completion, we stated the deadline for survey responses and sent regular email reminders about survey closure. We also contacted participants electronically to obtain clarification on incomplete or partially complete responses. We specified that the medical director or

Mahajan et al.
Global Survey of ED Responses to COVID-19 emergency preparedness expert in the division/department should be the one to answer this questionnaire.

Statistical Analysis
We tabulated descriptive statistics including absolute and relative frequencies for categorical variables and means (with standard deviations [SD]) or medians (with interquartile ranges [IQR]) for continuous variables, depending on normality, to compare all survey question responses. We defined ED characteristics by location, setting, and size. We grouped EDs by location into geographical categories: Asia; Europe; North America; South America; Africa; and Australia. Hospital setting was defined along two domains. First, we defined them as "public" or "government-funded" hospitals with the rest defined as "private" hospitals. Second, we categorized hospitals as "academic" vs "non-academic" based on presence of residency training programs. We used SAS 9.4 (SAS Institute, Cary, NC) for all quantitative analyses. ED characteristics stratified by country, continent, and income status (gross domestic product) as well as COVID-19 prevalence and death rates at time of survey completion are provided in Supplementary Table 1.

Pandemic Preparedness
Most sites (71/74, 96%) reported having an ED protocol to guide the screening, testing, and managing of suspected COVID-19 cases, and 27 (36%) sites sent their protocols for other EDs across the globe to use. There were 43/74 EDs (58%) that had ED staffing back-up plans prior to the COVID-19 outbreak. Regarding supply availability, 59/71 (83%) reported stocking the ED with personal protective equipment (PPE) as a part of pre-existing disaster plans. However, when respondents answered questions regarding effectiveness of existing disaster plans only 59/74 (80%) replied. Of these, 48/59 (81%) responded that the plan was successful/effective. Regarding personnel and ED staffing back-up plans, 26 felt they were very effective, 33 somewhat effective, six not effective, and nine did not respond. Reasons for disaster plans being successful or not are described in Table 1.

COVID-19 Pandemic Response
Nearly every ED (71/73, 97%) developed and implemented a protocol for screening, testing, and managing patients with suspected SARS-CoV-2 infections. The WHO recommendations for COVID-19 informed the pandemic response plans for 28/73 (38%) EDs in our sample. Outside the United States, 13/59 (22%) EDs based their pandemic response plans on the US Centers for Disease Control and Prevention (CDC) guidelines. The majority (45/73, 62%) based their plans on guidelines issued by their own countries.

ED Triage and Capacity
Screening criteria for COVID-19 were generally similar across the EDs in our survey, with less similarity across survey respondents in approaches for increasing treatment capacity. Screening criteria used are provided in Table 2 with the most common being fever (72/74, 97%), close contact with a confirmed case of COVID-19 (66/74, 89%), travel to a COVID-19 affected area (66/74, 89%), symptoms of upper respiratory illness (63/74, 85%), and signs of lower respiratory illness (71/74, 96%). However, 71/74 (96%) EDs reported that triage screening criteria had changed over time, and 58/74 (78%) mentioned criteria had changed more than two times. Of those responding, 68/74 (92%) created separate COVID-19 areas in ED waiting rooms, and 51/72 EDs (71%) increased capacity in response to COVID-19 surges, with 46 enhancing existing ED space by modifying/adding hallway beds, chairs/recliners or creating spaces separated by curtains. The results show that 45/74 (61%) increased capacity outside the ED by using non-traditional space for ED care such as subspecialty clinics or mobile tents. The most common measures used to increase hospital capacity were postponing elective/non-urgent surgical procedures (67/74, 91%), creating a dedicated COVID-19 patient care team (57/74, Global Survey of ED Responses to COVID- 19 Mahajan et al.  response. Most respondents (52/74, 70%) developed new or separate staff backup plans specifically for COVID-19 while 22/74 EDs (30%) had activated their existing staff backup plans at the time the survey was completed. From the received responses 47/74 (64%) reported staff missing work due to illness, and 32/74 (43%) reported that ED provider staff had tested positive for SARS-CoV-2. Inability to work due to possible COVID-19 illness was reported in nurses in 26/32 (81%) responses, followed by physicians, 21/32 (66%), and residents, 20/32 (63%). Common measures to address staff wellness and prevent burnout included allowing staff to work remotely (57/72, 79%) and providing meals at work (45/72, 63%). If staff had a confirmed positive SARS-CoV-2 test, criteria for returning to work included a negative test (28/45, 62%) and/or being symptom-free for a site-determined duration of days (27/45, 60%).

Testing for COVID-19
On a multiple-option question, 55/74 (74%) respondents selected that they were able to test for SARS-CoV-2 in their EDs, 23/74 (31%) selected that samples were sent to an external agency for testing, and 4/74 (5%) sites indicated that they did not perform any testing.

Resources and Supplies
Most EDs were able to provide face masks for patients with suspected COVID-19 or influenza-like illness at arrival

Innovation
In an open-ended survey section, we recorded both successful and unsuccessful innovations developed in respondent EDs to address the pandemic (Table 3). Successful innovations generally included development of separate treatment spaces for COVID-19 patients and plans for local community engagement via electronic and social media. Respondents reported unsuccessful innovations related to testing capacity, PPE availability, and staffing plans.

DISCUSSION
In this global survey, we leveraged the EMERGE network to establish rapid collaborations with non-EMERGE EDs and obtained estimates of ED pandemic preparedness and response to COVID-19 from 74 EDs in 28 countries comprising the six WHO regions. Despite substantial differences among surveyed EDs, there were many similarities in how EDs are responding to the pandemic, especially with screening protocols, capacity expansion, and staffing. Despite having disaster plans in place, globally ED pandemic preparedness and response is difficult and variable. There was a substantial human cost to the pandemic as 43% of the EDs reported providers and staff had contracted COVID-19 and missed work, further impacting the ED's ability to provide care. Finally, EDs were willing to share protocols, lessons learned, and innovative solutions that can be rapidly disseminated via research networks such as EMERGE, or via global organizations such as the WHO to benefit emergency care globally.
Our survey identified multiple, latent patient-safety threats due to inadequate disaster preparedness. One third of respondents did not have a pandemic preparedness plan prior to the COVID-19 pandemic. These findings are consistent with another report surveying 102 pediatric EDs in Europe early in the COVID-19 pandemic, which revealed that nearly a third of EDs lacked a contingency plan for pandemics and never had simulated scenarios for such events. 15 Second, the loss of the ED workforce, especially nursing and physician staff, either due to contracting infections with SARS-CoV-2 or required quarantine from exposure to patients with COVID-19 was very high. These results are consistent with reports from the CDC and other recent studies. [16][17][18][19] Third, there was variability among EDs regarding staff backup plans. Although most EDs responded that their pre-pandemic disaster plan was a successful one, nearly half of the EDs did not have an ED backup plan for staffing during disasters, most had to create new or modify existing plans to respond to the current pandemic and reported that these plans were ineffective or only somewhat effective.
Reassuringly, most EDs recognized the impact of the pandemic on staff wellness and had developed plans for improving provider well-being. These interventions included new guidelines for remote work whenever possible, meals during shifts, childcare support, and/or additional time away from work. Disaster management experts voice the importance of frontline workers' protection through planning, availability of PPE, and mindful staff scheduling, as well as strongly encouraging mental health and peer support with wellness initiatives. Our findings highlight the need for a more cohesive and comprehensive evaluation of existing disaster plans. Lessons learned from our survey could potentially be used to develop multidisciplinary in situ simulations to find optimal solutions for adapting to COVID-19 and other highly complex and evolving epidemics/pandemics in the future. 20 Nearly all EDs developed site-specific protocols for screening and triage during the pandemic. These protocols included provision of masks and separating patients on arrival to the waiting room. Triage and pre-ED arrival screening involved tele-triage, phone, and video screening, the use of alternative sites (such as triage tents, triage in the car), and use of other non-ED sites (such as outpatient clinics and community centers). One site reported a separate screening facility with app-based screening prior to patient registration. Leveraging technology can reduce ED demand and enhance patient understanding of COVID-19 risks and has been used extensively for screening and contact tracing tools, such as the CDC Self-Checker and Singapore's TraceTogether. 21,22 Consistent with lessons learned from prior disasters, most EDs had developed protocols for enhancing capacity by repurposing unused or non-ED spaces and adapting them for care of COVID-19 patients. Essentially, most EDs and their hospitals had re-engineered the entire ED arrival, triage, throughput, and disposition processes with a system-wide response, postponing non-emergent surgeries and creating and/or expanding intensive care capabilities.
The ED and emergency medical services are an integral component of community-based health systems that require Global Survey of ED Responses to COVID- 19 Mahajan et al.  strengthening to respond to pandemics; however, it is imperative that we first obtain high quality, global data in a timely manner to understand the impact and subsequent responses. This was the motivation for our survey, which revealed important insights on the state of global pandemic/ disaster preparedness. First, apart from regional EM organizations gathering local data, currently there is no infrastructure for assessing the state of ED systems across the globe and sharing experiences to provide rapid, reliable, and actionable data. For instance, adaptation of shared protocols and data would substantially reduce the time to implementation of interventions, underscoring the importance of ED-based research networks. Our survey was initiated in the nascent EMERGE network and we immediately recognized that despite having a "global" footprint, EMERGE is still not representative of the global ED system. It was very reassuring that we were quickly able to reach out to several non-EMERGE EDs via referral, allowing us to expand the survey participant pool from 26 EMERGE EDs in 17 countries to include an additional 51 non-EMERGE EDs from 11 countries. Second, our higherthan-average 57% response rate highlights the willingness of healthcare institutions across the world to participate and share experiences for the global benefit. 23 Third, the diversity of comments regarding innovations in the domains of triage, testing, communication, staffing, and capacity-building provided in Table 3 can be reviewed and potentially applied immediately as the world continues to grapple with the pandemic. For instance, use of unmanned robots for telepresence, use of mobile technology for communication between patients, caregivers, and providers, or use of a dedicated website and app-based screening have been rapidly deployed. Repurposing used radiographic films as face shields to three-dimensional printing of face shields, ultraviolet lights for sterilizing PPE to conserve supplies, negative pressure single-person hoods and tents to overcome lack of negative pressure rooms, and dedicated airway teams composed of emergency, critical care and anesthesia providers are all examples of local innovations that can be applied globally. Fourth, comments on whether pandemic plans were successful or not provide insight into how EDs, institutions, communities, and others can learn. For instance, sharing protocols and processes for COVID-19, such as airway management guidelines aimed at first-pass success with use of advanced airway placement devices (video laryngoscopes), and higher doses of paralytics to reduce risk of aerosol exposure, can be implemented globally. Sharing of ED mitigation strategies early in the pandemic will enhance cross-pollination of ideas that can promote both patient and healthcare worker safety. Conversely, evidence of inconsistent communication between providers within an institution or from regulatory agencies, inadequate PPE supplies, inability to scale testing to meet demand, and wasted PPE from improper training are examples of how plans could be improved in the future.

LIMITATIONS
Data obtained from surveys is inherently susceptible to selection and reporting bias. Furthermore, survey participation and response rates have declined over time due to limitations of the study design and inconvenience associated with completing poorly designed or frequent requests. This may be especially true during this pandemic, when multiple and often simultaneous requests for information were circulating broadly and a massive increase in scientific literature submissions related to COVID-19 were published. 24 We mitigated some of these limitations by testing the survey for face, construct, and content validity before soliciting participation. In addition, we made it very convenient for expeditious completion by formatting most questions as closed-ended responses. We obtained a 57% response rate, which is better than most reported surveys. 23 We also recognize that the two open-ended questions regarding effectiveness of existing disaster plans and interventions gave us rich anecdotal data on how individual EDs were responding, but we do not have details regarding their success/failure and/or impact. We plan to pursue these questions with follow-up interviews with ED leadership as our subsequent research project.
Given the extremely dynamic nature of the COVID-19 pandemic and the fact that countries were experiencing different levels of disease burden and different phases of the outbreak, the survey results represent the situation at a single point in time for the responding institutions. Thus, responses to survey questions will likely be different if we were to conduct the survey at another point in time. For instance, responses regarding availability of PPE or effectiveness of preparedness plans may differ as institutions continue to respond and adapt in real time. Ultimately, this study is hypothesis-generating and intends to demonstrate the power of our network.

CONCLUSION
Our study cohort represents a large cross-sectional sample of ED responses to an ongoing global healthcare crisis. Despite having disaster plans in place, ED pandemic preparedness and response continue to be a challenge and there were multiple, latent safety threats to providers and patients. We believe global emergency research networks play an important role in near realtime collection of high-quality data on the epidemiology of largescale events and can disseminate experiences and solutions that will impact healthcare outcomes for individuals, communities, and even nations.