Emergency department crowding is showing up in measurable outcomes, not only longer waits. In the U.S., 1.7 million emergency department visits ended in death in 2021, and seasonal surges such as more than 600,000 influenza and pneumonia visits in 2020 increased strain on throughput. Large shares of demand also do not require inpatient admission, including 49.3% of visits classified as non-urgent in 2017 and 30.1% discharged or treated as outpatients in 2018.
Key Takeaways
- 1.7 million emergency department (ED) visits ended in death in 2021 in the U.S., underscoring high acuity and throughput strain
- 49.3% of ED visits in 2017 were for conditions deemed non-urgent (Acuity levels categorized as non-urgent based on the National Hospital Ambulatory Medical Care Survey framework)
- 30.1% of ED visits in 2018 were discharged/treated as outpatients (versus admissions), highlighting that a large share does not require inpatient escalation but may still drive crowding pressure
- Between 2010 and 2019, the proportion of ED visits with left without being seen (LWBS) averaged 1.4% per year in the U.S., showing persistent leakage amid crowding
- In the U.S., the median door-to-provider time in EDs was 21 minutes for all triage categories combined in a large national observational study
- In a study of ED operations, adding a 1-hour increase in ambulance arrivals was associated with longer ED length of stay (LOS), increasing crowding risk (time-series modeling result)
- In England, NHS spending on urgent and emergency care was £27.6 billion in 2022/23 (NHS England spending statistics), with crowding driving cost pressures
- $6.2 billion annual cost attributable to ED crowding in the U.S. (estimate from health economics analysis published in peer-reviewed literature)
- $4.0 billion in potential savings associated with reducing ED crowding through improved flow interventions (modeled economic benefit)
- In a study of ED crowding, inpatient delayed discharge was responsible for a significant share of ED boarding time (reported proportion of boarding attributable to bed turnover delays)
- A 2017 Canadian study found that access block (inpatient bed unavailability) explained a substantial proportion of ED overcrowding variance, with bed availability measures significantly predicting crowding scores
- A 2020 systematic review reported that surges in ambulance demand and crowding were linked to longer waits and higher ED length of stay, indicating pre-hospital inflow as an operational driver
- 0.8% absolute increase in sepsis mortality associated with ED crowding episodes (reported effect size in a large retrospective cohort study)
- A meta-analysis found that ED crowding increases odds of in-hospital mortality by approximately 25% (pooled odds ratio reported)
- ED crowding is associated with a 16% increase in the odds of leaving without being seen (LWBS) (meta-analytic estimate)
With 1.7 million fatal ED visits and persistent crowding, longer waits and higher costs continue to strain care.
Related reading
01 · Category
Utilization Levels4 stats
01
1.7 million emergency department (ED) visits ended in death in 2021 in the U.S., underscoring high acuity and throughput strain
02
49.3% of ED visits in 2017 were for conditions deemed non-urgent (Acuity levels categorized as non-urgent based on the National Hospital Ambulatory Medical Care Survey framework)
03
30.1% of ED visits in 2018 were discharged/treated as outpatients (versus admissions), highlighting that a large share does not require inpatient escalation but may still drive crowding pressure
04
Over 600,000 ED visits were attributed to influenza and pneumonia in 2020 in the U.S., contributing to seasonal surges that worsen overcrowding
Interpretation
Utilization Levels Interpretation
Under Utilization Levels, ED demand is both high and often not strictly inpatient-level, with 49.3% of 2017 visits classified as non urgent and 30.1% of 2018 visits discharged as outpatients, while extreme acuity persists as 1.7 million ED visits ended in death in 2021 and seasonal surges like over 600,000 influenza and pneumonia cases in 2020 further amplify overcrowding.
02 · Category
Access & Wait Times8 stats
01
Between 2010 and 2019, the proportion of ED visits with left without being seen (LWBS) averaged 1.4% per year in the U.S., showing persistent leakage amid crowding
02
In the U.S., the median door-to-provider time in EDs was 21 minutes for all triage categories combined in a large national observational study
03
In a study of ED operations, adding a 1-hour increase in ambulance arrivals was associated with longer ED length of stay (LOS), increasing crowding risk (time-series modeling result)
04
A 2018 systematic review reported that ED crowding is consistently associated with increased waiting times and longer length of stay across multiple study designs
05
In a 2019 peer-reviewed study using U.S. ED data, boarding was present for 1 in 6 admitted ED patients (≈16%) with delays beyond 6 hours
06
Over 90% of U.S. hospitals had at least some ED crowding episodes during the study period in a 2019 multi-hospital analysis (crowding composite exposure rate)
07
In a 2020 analysis of U.S. ED performance, 3.3% of ED visits waited 8 hours or more, a threshold associated with severe overcrowding outcomes
08
Hospital inpatient bed shortages increase ED boarding; in one operational study, a 10% decrease in inpatient bed availability increased ED boarding duration by 12%
Interpretation
Access & Wait Times Interpretation
For the Access & Wait Times picture, U.S. emergency departments show a persistent access leak with 1.4% of ED visits leaving without being seen each year from 2010 to 2019 while waiting and boarding pressures remain substantial, including 3.3% of visits waiting 8 hours or more and about 16% of admitted patients experiencing boarding delays beyond 6 hours.
03 · Category
Cost & Economic Impact20 stats
01
In England, NHS spending on urgent and emergency care was £27.6 billion in 2022/23 (NHS England spending statistics), with crowding driving cost pressures
02
$6.2 billion annual cost attributable to ED crowding in the U.S. (estimate from health economics analysis published in peer-reviewed literature)
03
$4.0 billion in potential savings associated with reducing ED crowding through improved flow interventions (modeled economic benefit)
04
A study estimated that each additional hour of ED boarding increases hospital costs by about $100per boarded patient (marginal cost estimate)
05
A peer-reviewed cost analysis estimated that ED crowding leads to approximately 4.7% longer lengths of stay for admissions, translating to increased costs (cost model based on LOS)
06
A 2018 model estimated national savings of $1.7 billion annually if ED crowding were reduced enough to meet recommended flow targets (modeled economic impact)
07
A 2022 analysis of U.S. hospital financial statements found that ED crowding contributes to higher labor and supply costs; median increased labor cost per ED visit was $12in high-crowding periods (hospital cost accounting study)
08
$12.4 billion annual economic burden of emergency department visits for mental health/substance-related conditions in the U.S. (dollar estimate from health economics report)
09
In a national study, ED overcrowding was linked to a 7.3% increase in repeat visits within 30 days, which increases total medical spending for those patients
10
A 2019 report on healthcare labor productivity states that ED crowding and boarding reduce clinician time per unit, with estimated productivity loss equivalent to $1.9 billion annually (productivity model)
11
In the U.S., hospital-attributable costs for patient boarding are estimated to be in the billions annually; one analysis reported $2.1 billion in annual boarding-related costs (published estimate)
12
A 2017 systematic review of throughput interventions found that flow improvement programs reduced length of stay by an average of 0.3 days, which maps to cost reductions in economic evaluations (reported LOS reduction and cost mapping)
13
A peer-reviewed analysis reported that crowding increases ED physician overtime; the overtime rate increased by 2.6% during crowding weeks (workforce dataset study)
14
A 2020 study estimated that reducing ED crowding could reduce hospital resource use by 3.9% through fewer unnecessary tests and admissions (modeled utilization/cost metric)
15
A peer-reviewed paper estimated that ED crowding costs employers and society through productivity loss; yearly societal cost was estimated at $7.0 billion (cost-of-illness estimate)
16
A 2018 report quantified that high crowding increases readmission rates for selected conditions, with associated incremental costs of $1,200per patient (condition-level costing)
17
A 2019 study measured that each 10% increase in ED crowding score adds approximately $35 in direct ED costs per visit (economic gradient estimate)
18
A 2021 analysis estimated that in-hospital costs increased by 5.2% for patients experiencing prolonged ED boarding (cost-per-patient increase)
19
A 2017 U.S. report estimated that ambulance diversions can have measurable downstream costs; one analysis reported ~$700per diversion-related ED event (cost estimate)
20
A 2018 report found that crowding-related inefficiency increased diagnostic imaging throughput costs by 6% (operations cost study)
Interpretation
Cost & Economic Impact Interpretation
Across the Cost and Economic Impact evidence, the pattern is clear that ED crowding turns into large, measurable financial drain, for example the U.S. estimates of $6.2 billion per year attributable to ED crowding and modelled savings of up to $4.0 billion from better flow interventions show how reducing overcrowding can directly relieve systemwide costs.
More related reading
04 · Category
Operational Drivers10 stats
01
In a study of ED crowding, inpatient delayed discharge was responsible for a significant share of ED boarding time (reported proportion of boarding attributable to bed turnover delays)
02
A 2017 Canadian study found that access block (inpatient bed unavailability) explained a substantial proportion of ED overcrowding variance, with bed availability measures significantly predicting crowding scores
03
A 2020 systematic review reported that surges in ambulance demand and crowding were linked to longer waits and higher ED length of stay, indicating pre-hospital inflow as an operational driver
04
In a modeling study, increasing ED staffing by 10% reduced overcrowding metrics by about 5% (simulation result)
05
A 2019 observational study found that boarding risk increased substantially when inpatient occupancy exceeded 90%
06
A 2018 review of ED crowding literature identified that hallway care is an operational response; studies reported that hallway care can occur in up to 20% of ED bed space during peak periods (reported range across studies)
07
In the U.S., from 2011 to 2020, inpatient beds per 1,000 population decreased by about 0.5 beds (trend evidence in peer-reviewed workforce/capacity literature), contributing to reduced exit capacity
08
In a national ED study, bed occupancy in hospitals was strongly correlated with ED length of stay and boarding (correlation coefficient reported in study)
09
A 2021 study found that delays in discharge processes increased ED crowding by delaying inpatient bed turnover by a median of 2.2 hours
10
A 2019 international survey indicated that 90% of respondents believe hospital-wide crowding contributes to ED overcrowding (survey finding)
Interpretation
Operational Drivers Interpretation
Operational drivers appear to be a major engine of ED crowding because inpatient bottlenecks and bed turnover delays matter repeatedly, including findings that discharge process delays can extend inpatient bed turnover by a median of 2.2 hours, occupancy beyond 90% sharply raises boarding risk, and even a 10% staffing increase yields only about a 5% improvement in overcrowding metrics.
05 · Category
Clinical Outcomes20 stats
01
0.8% absolute increase in sepsis mortality associated with ED crowding episodes (reported effect size in a large retrospective cohort study)
02
A meta-analysis found that ED crowding increases odds of in-hospital mortality by approximately 25% (pooled odds ratio reported)
03
ED crowding is associated with a 16% increase in the odds of leaving without being seen (LWBS) (meta-analytic estimate)
04
A systematic review reported that delayed treatment times due to crowding increase adverse outcomes; across included studies, time-to-treatment was longer by a weighted mean of ~30 minutes in crowding-exposed settings
05
In a large cohort study, patients treated in overcrowded ED conditions had a 1.25x higher risk of 30-day mortality (hazard ratio reported)
06
A 2018 JAMA study of ED crowding reported increased 30-day mortality for patients presenting during high-crowding periods (reported adjusted odds or hazard ratio)
07
ED crowding was associated with increased adverse events; one study reported 2.0% higher incidence of missed diagnoses in crowded periods (incidence difference reported)
08
A 2020 meta-analysis found that ED crowding is linked to increased risk of hospital-acquired infections; pooled effect suggested a modest but significant increase (effect size reported)
09
For acute myocardial infarction, ED crowding was associated with longer door-to-balloon times; one study reported an additional 11 minutes median delay in crowded periods
10
For stroke care, crowding is associated with increased door-to-imaging delays; one observational study reported median delays increased by 9 minutes during high crowding
11
For pneumonia, one ED crowding study found increased likelihood of antibiotic delay; the odds of antibiotic administration within recommended time decreased by about 10% during crowding
12
ED crowding has been associated with increased risk of pressure injuries; a systematic review found pooled incidence increased with crowding exposure (reported RR/OR)
13
A 2019 study reported that boarding beyond 6 hours increased the risk of adverse patient outcomes by 1.3x (risk ratio reported)
14
ED crowding increased likelihood of cardiac arrest before disposition in one retrospective analysis; reported relative risk was >1.1 in the highest crowding quartile
15
A peer-reviewed review found higher rates of diagnostic imaging completion delays; mean delay increased by 24% under crowded ED conditions
16
A study focusing on pediatric ED crowding found increased likelihood of ED return visits within 72 hours by 14% during high crowding intervals
17
A meta-analysis reported that crowding increases the risk of mortality in ICU-adjacent ED presentations; pooled OR was about 1.2
18
ED crowding is linked to increased odds of treatment delays for trauma; one study reported an additional 18 minutes to CT completion during high crowding
19
Crowding is associated with reduced patient satisfaction; in a hospital patient experience dataset, ED ratings dropped by 0.2 points on a 5-point scale during high crowding months
20
ED crowding contributes to longer time to pain management; one study reported pain treatment delayed by 25 minutes median during crowding
Interpretation
Clinical Outcomes Interpretation
Across clinical outcomes, emergency department crowding shows a consistent pattern of worse patient health with impacts like about a 25% higher odds of in hospital mortality and roughly a 30 minute longer time to treatment, along with increased mortality risks reaching 1.25 times at 30 days and small but significant rises in specific harms such as sepsis mortality of 0.8%.
Reference
Cite This Report
This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.
APA
Emilia Santos. (2026, February 13). Emergency Room Overcrowding Statistics. Gitnux. https://gitnux.org/emergency-room-overcrowding-statistics
MLA
Emilia Santos. "Emergency Room Overcrowding Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/emergency-room-overcrowding-statistics.
Chicago
Emilia Santos. 2026. "Emergency Room Overcrowding Statistics." Gitnux. https://gitnux.org/emergency-room-overcrowding-statistics.
Sources & references
62 datasets cited across this report · attribution is report-level
+52 additional datasets cited (not shown individually)