A conceptual framework for Emergency department design in a pandemic

Although individual ED experiences during infectious disease outbreaks have been described in the literature [16‐22], few works have outlined a pragmatic framework incorporating sound principles to guide ED design and workflow adaptations. This paper aims to address issues around ED design and workflow amidst the COVID-19 pandemic. We propose a conceptual framework and checklist for EDs to be ready for future outbreaks without compromising care of the broader patient population needing the ED.

Situational awareness and a surveillance system are paramount in identifying and isolating the first suspected cases and being alerted to new clusters. In a busy ED environment, early identification of an outbreak or cluster pattern is a major challenge. A good surveillance system needs to be in place even during non-outbreak periods.

Perimeter defence refers to the formation of a conceptual “perimeter” through ED design and workflow to segregate potentially infected patients from other non-suspect cases. This is a key containment measure as this protects the ED, hospital, staff and patients. Two spatial control strategies can be employed (defensive isolation versus offensive containment [27]) depending on the stage of the outbreak and the load of infectious suspects.

One of our top priorities is to advocate for zero infection amongst healthcare workers [23]. This can be done in several ways. First, there should be adequate PPE supply and appropriate training on their use. Risk-adapted use of PPE should be part of routine ED practice when managing any infectious diseases like chicken pox or tuberculosis. In an outbreak situation, supplies need to be scaled up rapidly to meet the expected increase in PPE requirement. Second, roster modifications such as adaptation of modular shift schedules [28] can allow for effective contact tracing and mitigate the risk of widespread nosocomial infections. Lastly, institutions should also have the capability to screen and test staff for illness and if available, provide them with antiviral prophylaxis and vaccines in view of the inherent risks they face in the line of duty.

During a pandemic, ED caseload surges are inevitable. Surge capacity is the ability to manage patients above the usual volume [29] without compromising normal care. Surge capacity can potentially be increased by managing demand, establishing alternate care facilities, expanding bed capacity and minimizing resource consumption by patients [30]. The system needs resilience and flexibility to remain functional within the capacity of its available resources during such outbreaks [26].

The ED needs to have a system in place to allow recovery to its usual steady state. This includes infrastructure as well as staff. Frontline staff need to be able to function at full capacity even after the outbreak period. At its recovery stage, ED volumes may even be higher due to the rebound of patients with complications from neglect of chronic conditions.

Incorporating the principles above, we propose the following conceptual framework to depict this (Fig. 1). The principles form the border of our conceptual framework as they should govern international guidelines, national and local health system operations, research, public and staff education, and eventually the operations within the hospital and ED. International guidelines, health systems, research and education form the background that support hospital and ED operations, making way for cost effective policy implementation.

For the principles to be effectively implemented in the hospitals and ED, there needs to be good leadership, supporting policies and an effective flow of communication down the command line; hence forming the apices of the triangle [31].

The operations of the hospital and the ED are intertwined and limited by their resources as defined by space, staff and supplies. During a crisis, the perimeters of this triangle may be expanded (hence depicted by dotted lines); but is dependent on institutional leadership, policies and communication capabilities (as depicted by the apices of the triangle).

Space, staff and supplies from the hospital can be diverted to the ED to deal with the surge initially. Surge patients also need to be able to flow to infectious diseases units and intensive care units (ICUs) of the hospital. The following section describes a pragmatic approach within the ED workflow using the model above.

Within the ED, patients should be seen and managed within their allocated areas (high, intermediate and non-suspect areas). There should be physical barriers between these allocated areas. Ideally the high suspect cases should be managed in isolated, negative pressure cubicles. The intermediate suspect can be managed as a cohort in a space which allows adequate (e.g.: 1 m) spacing between them. The use of radio-frequency identification (RFID) tags may be useful to allow tracking of the patients’ movements.

Visitor movements should be restricted, and policies made restricting visitation within the high or intermediate suspect areas. For this to occur, security should be activated and the ED entrances and exit areas should be cordoned off.

Staff working in a specific designated area for infected patients should ideally remain in that space throughout a work shift. If there is a need to travel to a different section, the full PPE should be changed. There should be clear instructions on the type of PPE to be worn in the respective areas. Disposable scrub sets and showering facilities within the ED are also important for staff protection.

In some systems, a modular staffing approach is adopted, such that one physician and nurse team remains in the high-risk suspect area for a specified period. Another type of modular staffing is a team-based, fixed shift system where the physicians and nurses work within the same team during the outbreak period to limit contact within the department [28]. If a team member is infected, the rest of the team members can be quarantined. This will also facilitate contact tracing when positive patients are seen by the team.

Aerosol generating procedures will need to be modified in a situation such as COVID-19, including reducing the use of nebulisers, non-invasive ventilation and even high flow oxygen. Appropriate equipment, PPE such as Powered Air-Purifying Respirator (PAPR) and a defined protocol needs to be in place for intubation procedures such that the staff involved are protected.

Within the three areas, there must be the capacity to manage patients of all clinical severity levels. Each area requires workflows for time sensitive conditions such as trauma, myocardial infarction or stroke. Relevant activations should occur for these cases, and the receiving specialist teams need to aware of the infectious profile of the patients (eg: high or intermediate suspect) so that the appropriate, timely disposition can be arranged.

Patients with suspicious complaints but who do not fall under the case definition criteria may have to be placed in the intermediate suspect area [41]. This will ensure that at the initial stages, when the disease transmission profile is unclear and the situation is dynamic, patients who could be potential suspects are placed away from non-suspects.

Consideration needs to be given to palliative or dying patients [21]. Many of them will present with respiratory symptoms. If they are seen in an intermediate or high-risk area, there should be a workflow to determine the number of family members allowed by their bedside and if possible, a separate, private room.

Some investigations will need to be expedited to reduce throughput times, for example swab results or chest x-ray reports. This will require a collaborative workflow with radiology and clinical laboratory departments. The ED can also limit and stratify relevant investigations to reduce strain on resources. The areas for investigations may need to be replicated for the high and low suspect areas such as availability of plain radiography capabilities in the high suspect areas.

Emergency observation units may have to cease operations or modified to make space and human resources available for the surge in patients.

If the investigations are not able to yield results during their stay in the ED, the stratification of high, intermediate and non-suspect must be maintained throughout, even at the output stage. In this situation, physical ward spaces should be in line with the ED suspect stratification system. For example, there may be a need for three different destination wards for the patients (e.g.: high suspect, intermediate suspect, non-suspect).

There should ideally be rapid movement of patients up to the ward. Hence, there needs to be close liaison with the inpatient wards and bed management unit (BMU). Delays are still to be expected and there should be an observation or in-patient bed waiting area within each of the suspect areas.

There should be workflows to help decide those who are safe to be swabbed and discharged. The ED can be designed such that those who are well enough to be swabbed and sent home, may not even need to enter the actual ED, as seen in Taiwan [42].

For those discharged, it is important to determine if they need to be on a stay home notice (SHN), and there should be an approach to perform contact tracing. Appropriate return advice should be given. The ED can work closely with the public health to enforce SHN by various methods (e.g.: RFID tagging, telemedicine).

After patient disposition is decided, patients may be moved to other areas to await transfer to beds or other isolation facilities. Proper hand-over between staff is essential. Designated, risk-adapted waiting areas also need to be available to maintain segregation of high suspect cases. Transfer of high suspect patients out of the ED itself can be challenging, as there is a risk of transmission en-route. In order to protect staff and passers-by from inadvertent exposure, the pathway for patient movement needs to be cleared, a task that may involve activating hospital security. If transfers to another facility is required, designated ambulance services should be available where transferring staff are aware of the risks and have proper PPE. Frontline staff protection also includes assigning dedicated decontamination (eg: showers), rest and recuperating areas. Having rest areas without a proper contamination area can be counterintuitive towards staff protection. If there are space constraints for a staff decontamination area, an alternative is to have a ‘clean’ and ‘dirty’ staff rest area. The system should also have a staff wellness or psychological first aid program in place to protect them from mental fatigue or burnout which are potential issues of such a pandemic.

There needs to be a set of criteria to determine if patients require admission and the type of ward required (individual isolation for high suspect, cohort ward for intermediate suspect or normal ward for non-suspects). Close communication with the BMU is essential to ensure that patients are cleared quickly from the ED to accommodate for surges. Cancellation of elective admissions and expedited discharge of patients may need to take place to allow the flow of patients to the wards.

From the key concepts identified above, we created a workflow model (Fig. 2) and checklist (Table 1) to guide ED design and workflow during COVID-19 as well as future outbreaks.

Infectious disease outbreaks pose specific problems to the ED. These problems can be divided into four timeframes as depicted in Fig. 3: pre- outbreak, peak, plateau and post outbreak stages [31]. Before an outbreak, the main issue is with identifying possible index cases seeking treatment in the ED. What should the workflow encompass such that the net for suspects is cast wide without placing a strain on resources? What PPE should the staff at the frontline wear? How should patients enter the ED to mitigate the risk of spreading infection to staff and other patients?

At the peak of the outbreak, there are multiple challenges [43] such as suspect case identification, demand surge, staff safety, patient safety as well as the usual ED workload. The screening of patients, if performed variably, can have hazardous downstream effects such as hospital outbreaks. There needs to be a balance between under and over-triage, keeping in mind local and national guidelines and hospital resources, all of which are dynamic. Not all patients will have typical symptoms, adding to the challenges of designing the ‘perfect’ screening protocol. Availability and performance of test kits may also be an issue, especially early in an epidemic. There can also be difficulties in reconciling public health guidance with the practicalities of ED practice [15]. At the plateau stages, surge capacity will pose a major problem, both at the ED and the hospital level.

Figure 3 also depicts the conceptual change in ED workload over time. As the infectious disease workload increases, the usual ED workload may decrease for various reasons. When the outbreak settles and infectious case load decreases, there will be a spike in the normal ED workload. Understanding the ED trend in the local context will help guide the system, space, staff and supplies.

After an outbreak, there will be a surge of acute exacerbations of chronic illnesses, following lockdowns, physical and social isolation and the cancellation of clinic appointments and semi-elective surgeries. The ED should have a system in place to allow recovery to its usual steady state. This includes infrastructure as well as staff. Frontline staff need to be able to function at full capacity even after the outbreak period. At its recovery stage, ED volumes may even be higher due to rebound of patients with complications from neglect of chronic conditions or care-delays that were inevitable during times of crisis.

Preparation is key to mitigate the disastrous effects of such a pandemic and this preparation needs to be at a National, Hospital and Departmental level. COVID-19 has highlighted the lack of this preparedness can impact lives, strain resources and economic losses. At the frontline, the ED is the defence fortress of the healthcare system. It needs to be in a constant state of preparedness which can be adapted to various pathogens (eg: supply chain of appropriate PPE for highly transmissible pathogens like Ebola).

This is an area which warrants further research and guideline development. Based on previous outbreak experiences and the local health system, the public may rush to the ED to be further investigated during an outbreak. If the virus has a high transmission rate with low fatality, perhaps a community approach [20] to screening may reduce the hospital load, allowing them to focus on the critically ill.

In many parts of the world, Emergency Medicine is an under-recognised and under-resourced specialty. This may make the implementation of such a framework challenging. However, the ED or its equivalent is the gateway to the hospital and hence needs to be well established. Alternatively, this framework can be implemented within a department which functions as an ECS.

Another challenge is the cost of applying such a framework. Hence our deliberate attempt to propose a principle-based framework rather than prescribe an actual workflow for other countries. Future work can look at the level of implementation of the checklist with country specific examples.