Emergency Departments: The Queue at the Front Door Can Be Caused by a Blocked Exit
Emergency-department crowding is not always an emergency-department capacity problem. Boarding begins after a patient has been admitted but no inpatient bed is available, so the visible queue can sit in the ED while the binding constraint lies farther downstream in inpatient capacity, discharge readiness, actual discharge or community support. The systems task is therefore to follow patient flow end to end and locate the capacity that is failing to regenerate.
The Brief in One Sentence
A queue shows where congestion becomes visible; it does not necessarily show where the system constraint originates. In hospital flow, an admitted patient can wait in the emergency department because the downstream bed that should receive that patient has not yet become available.
Why It Matters
AHRQ’s 2025 work on emergency-department boarding makes the system boundary explicit: boarding occurs after the decision to admit when no inpatient bed is available, and its causes originate at hospital or health-system level rather than inside the ED alone.
NHS England reaches a similar operating conclusion from another health system. Its 2025/26 urgent-and-emergency-care plan says that in some trusts one in four bed days are lost to delayed discharge. Its 2026 Model Discharge Pathway states that the whole urgent and emergency pathway depends on patients leaving hospital when ready and links delayed discharge with prolonged waits for admission in emergency departments and corridor care.
These sources describe different countries and institutions, so their magnitudes should not be combined. Their useful convergence is narrower: patient flow is a circulation problem across the hospital and its interfaces, not a queue owned by one department.
Explain It Simply
Imagine a small restaurant with ten tables. The kitchen becomes faster and can prepare meals twice as quickly. But customers who have finished eating cannot leave because the payment desk is blocked.
The entrance queue grows. Adding another cook may make the kitchen even faster, but it does not free a table.
A hospital is far more complex, but the systems lesson is similar. If admitted patients cannot move to inpatient beds because those beds have not been released, local ED speed can improve while boarding remains.
System Boundary
This brief follows an urgent patient from arrival through ED assessment and treatment, decision to admit, inpatient-bed assignment, ward care, clinical discharge readiness, actual discharge and release of the bed for the next patient. It also includes the interfaces with diagnostics, pharmacy, transport, community care, social care and mental-health services when they determine whether flow can continue.
It does not claim that every ED delay is caused by discharge. Arrival demand, acuity, staffing, diagnostics, internal ED design, behavioral-health capacity, ambulance handover and other constraints can dominate in specific settings.
The analytical question is narrower: when admitted patients are boarding, what downstream event must occur before capacity becomes available again?
Evidence Map
- Observed / AHRQ 2025: ED boarding occurs after the decision to admit when an inpatient bed is unavailable; AHRQ says the causes arise at hospital or health-system level and require solutions beyond the ED walls.
- Observed / NHS England 2025/26: in some trusts, one in four bed days are lost because of delayed discharge; the plan requires a whole-system approach involving acute trusts, local authorities and integrated care boards.
- Observed / NHS England 2026: the Model Discharge Pathway says timely discharge is essential to flow across the urgent and emergency pathway and explicitly links delayed discharge with prolonged ED admission waits and corridor care.
- Observed / NHS England 2026 DRD guidance: Discharge Ready Date records when a patient no longer requires hospital-only care; actual discharge can occur later, making the delay after clinical readiness measurable.
- Observed / NHS acute-flow guidance: delayed patients in January 2025 had an average delay of 6.1 days, estimated at 324,000 bed days lost. This is England-specific operational evidence, not a universal benchmark.
- Inference: a bed occupied after clinical readiness is both a discharge delay for one patient and capacity not yet available to the next patient.
- Uncertain: the dominant cause of boarding varies by hospital, hour, patient group and local care system; event-level data are required before choosing a local intervention.
The Queue Is Upstream, the Constraint Can Be Downstream
Queues form where flow can no longer continue. That point is not necessarily the origin of the constraint.
An ED bed can remain occupied because the patient needs an inpatient bed. That inpatient bed can remain occupied because another patient is clinically ready but transport, medication, social support, community capacity or another discharge dependency is incomplete.
The visible chain can therefore be:
delayed discharge → bed not released → admitted patient cannot leave ED → ED capacity stays occupied → new arrivals accumulate.
This is a mechanism, not a claim that every crowded ED follows this chain.
Capacity Must Circulate
A hospital bed is not useful capacity merely because it exists physically. It becomes available capacity only after the current patient no longer needs it, the discharge actually occurs, the bed is turned over and the next patient can move.
This creates an important distinction between installed capacity and circulating capacity. A hospital can own the same number of beds while the effective capacity available to new admissions changes hour by hour depending on how quickly beds return to the pool.
Throughput therefore depends not only on how much capacity exists, but on how reliably it regenerates.
Actors and Incentives
- Emergency department: wants safe assessment, treatment and timely onward movement.
- Inpatient wards: manage current patients, admissions, treatment progress and discharge readiness.
- Bed-management teams: match demand with available beds across specialties and locations.
- Diagnostics and pharmacy: can accelerate or delay decisions and discharge when tests, results or medicines are pending.
- Community and social-care services: can determine whether a clinically ready patient can safely leave acute care.
- Transport and families: can become final-mile constraints even after the clinical decision is complete.
- Ambulance services: experience upstream consequences when ED capacity cannot accept new arrivals efficiently.
- Executives and system partners: must arbitrate across organisational boundaries where no single department owns the full flow.
Four Feedbacks to Watch
- Discharge delay → lower bed availability → boarding: fewer beds return to the pool, so admitted patients wait longer upstream.
- Long stay → harder recovery: NHS guidance notes that unnecessary hospital stays increase deconditioning, infection, delirium and dependence; some patients may therefore become harder to discharge safely the longer the unnecessary stay continues.
- Local speed → downstream queue shift: faster ED decisions can expose the inpatient constraint sooner if downstream capacity does not regenerate at the same pace. This is an inference, not a claim that faster ED care is harmful.
- Uneven discharge timing → recurring peaks: if releases are concentrated in limited hours or days while arrivals continue, capacity can tighten predictably even without a change in total bed count.
Sidy's Synthesis — Find the Capacity That Is Not Regenerating
A queue tells you where the system is suffering, not necessarily where to intervene.
My synthesis follows four questions:
- Where is the patient waiting?
- What event must happen next?
- What capacity must be released or created for that event?
- Why is that capacity not regenerating at the required rate?
This is my analytical extension, not an AHRQ or NHS named framework.
A patient who is ready to leave but still occupies a bed is not only a delayed discharge; that bed is also capacity missing from the next patient.
AI Can Improve Visibility — It Cannot Manufacture a Bed
AI and forecasting can make parts of hospital flow cheaper to see: expected arrivals, likely discharge dates, bed demand, patients at risk of delay, documentation gaps and mismatches between demand and available capacity.
What remains constrained is physical and institutional: staffed beds, clinical readiness, diagnostics, transport, community capacity, social care, mental-health capacity and the authority to resolve cross-organisational barriers.
New failure modes also appear. A model can create false precision around discharge timing, reproduce historical access bias, optimize a proxy rather than patient safety, or make a badly recorded workflow look mathematically clean.
Human judgment becomes more important in clinical admission and discharge decisions, exceptions, escalation and trade-offs between flow and safety. These decisions should not be delegated solely to a prediction model.
Where the Leverage Points Are
- Plan discharge from admission rather than when the bed is already needed.
- Separate clinical readiness from the operational tasks still preventing actual discharge.
- Profile bed release across the day and week, not only as a daily total.
- Measure admitted-patient boarding and delayed discharge together where the local pathway supports that linkage.
- Escalate cross-organisational delays to actors with authority to resolve them.
- Use same-day or ambulatory pathways where clinically appropriate to avoid unnecessary bed occupancy.
- Improve real-time visibility of the next event required for every waiting patient.
- Remeasure end-to-end flow after intervention so the queue is not merely shifted elsewhere.
What Would Reopen the Thesis
Reopen the diagnosis when event-level local data show that ED staffing, diagnostics or internal processing dominate the delay; when inpatient capacity expands but boarding does not improve; when discharge performance improves without changing boarding; when demand or acuity changes materially; when mental-health or another specialist pathway proves to be the binding constraint; or when a redesign changes where patients wait and which capacity is scarce.
Remember This
Do not fix the queue before you know which capacity stopped circulating.
Primary sources
Facts, figures and quotations should be traceable to the sources below. Sidy's synthesis is labeled as synthesis and does not replace sourced facts.
- AHRQ Report Identifies Strategies To Reduce Emergency Department Boarding — Agency for Healthcare Research and Quality
- Emergency Department — AHRQ topic and boarding resources — Agency for Healthcare Research and Quality
- Urgent and emergency care plan 2025/26 — NHS England
- The Model Emergency Department: high performing urgent and emergency care pathways — NHS England
- Model discharge pathway — NHS England
- Discharge Ready Date guidance and scenarios — NHS England
- Urgent and emergency care acute patient flow — NHS England