How to solve the problem of emergency department crowding
Emergency department crowding happens when the demand for emergency critical care exceeds available supply, and poses a serious threat to safe patient care.
What are the main causes of emergency department (ED) crowding?
Every emergency department is unique and crowding is a multifaceted problem, but a helpful framework does exist to categorise the three overarching sources of crowding.
Firstly, as the population grows, so too does the number of attendances to the ED. This crowding has also been further bolstered by public health campaigns focusing on time-critical conditions, such as heart attacks. In the UK, we’ve also seen a marked increase in ED visits post-COVID. So input causes of ED crowding are largely socio-economic.
There are then causes of crowding which are found once a patient has entered the ED. Bottlenecks are caused as facilities are often not equipped to deal with the increased demand due either to inadequate staffing levels or physical layout. Many other factors can cause a reduction in patient flow which leads to crowding such as the time of day, for example if hospital occupancy is highest in the evening then this is when patients in the ED will be waiting to gain access to beds still occupied by other patients.
Finally, reduced output from the ED can cause crowding; this is more commonly referred to as "exit block". A shortage of inpatient beds relative to demand can lead to increased length of stay for admitted patients in the ED. Exit block is often considered to be the most significant contributing factor. This is why ED crowding is a hospital-wide issue, and this must be considered in attempts to resolve it.
How does it impact patient care?
Crowding of EDs is the greatest threat to safe patient care in acute settings across the developed world. It impacts patient care in terms of the quality of care they receive and naturally this impacts patient outcomes.
ED crowding is proven to be associated with higher workload for staff, delayed patient assessment, higher costs of treatment, more frequent discharging of patients with high-risk clinical features, poor infection prevention and control measures, and lower patient satisfaction, which is also linked to lower likelihood of patient compliance with their follow-up plan.
This translates into lower patient outcomes, specifically in the form of high patient re-admission rates, prolonged hospitalisation, increased walkouts, a higher frequency of medication errors and adverse events, along with both increased morbidity and mortality.
It’s demoralising for staff as they cannot provide the level of care they aspire to when emergency departments run consistently with a demand that exceeds both the physical and staffing resources and capacity.
How has the pandemic impacted this issue?
During the pandemic, efforts were made to keep patients out of hospital wherever possible. This lowered ED admission rates, but as the world returns somewhat to normal, there has been a marked increase in attendance across UK EDs in the post-pandemic period, and the challenge has arisen once again.
Crowding has been further compounded by the need to appropriately space patients to deliver safe care in the context of the pandemic.
It has highlighted the need for solutions that are deliverable in the short to medium term as demand will only continue to increase, as has been the case over the last 20 years. And as the main pressure facing the ED, the requirement for ED leaders to actively engage in developing solutions to crowding is self-evident. ED corridors crowded with patients on trolleys and in chairs should not be an accepted part of 21st century healthcare.
Where is technology currently being used to tackle this problem, if at all?
Although the impact on patient care is well recognised, solutions to date are patchy and inconsistent.
In terms of technology, tools which provide superficial measurements of crowding to aid decision making such as the NEDOCS and ICMED scores, for instance, are available to emergency medicine leaders, though their limitations and shortcomings are widely agreed upon.
Otherwise initiatives tend to focus on increasing access to primary care and general practitioners, as well as alternative models of care. All have their flaws. The Royal College of Emergency Medicine has consistently argued that the proportion of low-acuity patients (who could be treated in alternative healthcare settings) is no more than 15%, so the effect of re-routing low-acuity and ambulatory patients to primary care clinics or increasing access to these clinics will likely be limited.
The UK has also introduced targets, such as for all ED patients to be treated within 4 hours. This has increased resources available to EDs, but has not kept pace with the year-on-year increase in demand.
Overall this is an area where technological innovation must be explored.
Can artificial intelligence and machine learning (AI/ML) help tackle the problem?
I believe only large datasets with AI/ML technology applied are capable of unlocking the proactive modelling that is required to combat the problem of overcrowding. In doing so, AI/ML offers the promise of transforming the provision of acute services from the current reactive system to a proactive model.
To achieve this, we need a testable prediction tool for both demand on the ED and admission to in-patient areas. This would be the first step in building a system that optimises available resources to meet anticipated pressures with a consequent reduction in ED crowding and the harm this causes.
Such a solution would transform the delivery of care in EDs and go a long way to ensuring safe and timely patient care while minimising clinician burnout.
What role will tech play in this area in the future?
In the future, I see the advent of truly advanced digital health technologies also playing a role in revealing extremely rich, previously inaccessible information on the real-time physiological health (vital signs etc.) of patients, as well as their exact location in the hospital. This may one day inform ML-driven crowding models with this kind of data also, which could transform our understanding of what constitutes quality hospital care management for good.
ED crowding is therefore an issue in the healthcare landscape where desperate need for change is meeting huge potential for innovation, which is why now is the time to form an international consortium to capitalise on this convergence.
You're involved with electronRx, what are they doing in this field?
electronRx is a deep tech startup based in Cambridge, UK. They have an expert team of interdisciplinary scientists and engineers developing a range of novel technologies to revolutionise patient engagement and inform clinical decision making, taking a staunchly data-driven approach to transforming how we deliver healthcare and treat disease.
With electronRx, we are building out an international consortium of Emergency Medicine leaders who are all passionate about collaborating to tackle the long-standing, internationally recognised barrier to delivering high-quality patient care that is ED crowding.
Our project looks to harness their AI/ML prowess to extract the value lying dormant in a wealth of previously inaccessible healthcare insights across the entire hospital. Our goal is to develop a holistic, AI-driven solution that provides actionable insights with measurable outcomes to tackle ED crowding once and for all.