Emergency Department Boarding Measures: The Metrics Hospitals Track

Emergency department (ED) boarding is the practice of holding a patient in the ED, on an ED gurney or in an ED hallway, after the decision has been made to admit them, because no inpatient bed is available. For a patient-safety officer, quality director, or risk manager, boarding is not a scheduling inconvenience — it is a documented risk factor: extended boarding is associated with higher inpatient mortality, longer overall length of stay, delayed medication administration, and a measurable rise in adverse events, which is why boarding time now sits alongside infection-prevention and medication-safety metrics on most hospital quality dashboards. This guide covers the specific metrics a boarding-measurement program tracks, how those metrics map onto the CMS ED throughput measures your quality team may already report on, and the Joint Commission expectation that ties boarding directly to a leadership accountability requirement.

What Counts as Boarding, Operationally

A patient is boarding from the moment an admission decision is documented (the “admit decision time”) until the moment they physically leave the ED for an inpatient bed, observation unit, operating room, or transfer. That start point matters for measurement integrity: counting from ED arrival instead of admit-decision time conflates boarding with the entirely separate problem of ED treatment-phase delay, and the two have different causes and different fixes. A clean boarding metric isolates the post-decision wait — the interval attributable to bed availability and placement logistics, not to diagnostic workup.

The American College of Emergency Physicians (ACEP) and the Institute for Healthcare Improvement (IHI) both define boarding this way in their respective flow and crowding guidance, and it is the definition CMS used when it built its own ED throughput measure set (below). Using a different start point makes a hospital’s numbers non-comparable to state or national benchmarks, so this is usually the first thing to check when a boarding metric looks unusually good or unusually bad against peers.

The Core Boarding-Time Metrics a Quality Program Tracks

Four metrics do most of the work in a boarding-measurement program. Each answers a different operational question, and a dashboard that reports only one of them (median boarding time, most commonly) will miss real problems the others catch.

1. Median admit-decision-to-inpatient-bed time

The core boarding metric: the interval from admit-decision timestamp to the timestamp the patient physically arrives on the inpatient unit (or leaves the ED for the OR, observation, or transfer). Reported as a median rather than a mean because boarding-time distributions are heavily right-skewed — a handful of extreme outliers during a capacity crisis will distort a mean badly enough to mask what a typical patient actually experiences. Most programs also track the 90th percentile alongside the median specifically to surface those outliers, since the median alone can look stable even while a worsening tail goes unnoticed.

2. Left-without-being-seen (LWBS) rate

The percentage of ED patients who register (or are triaged) and then leave before a provider evaluates them — a direct proxy for crowding severity and a patient-safety indicator in its own right, since a share of LWBS patients have conditions that needed urgent care they didn’t get. It’s calculated as LWBS encounters divided by total ED registrations for the period, and it should be tracked separately from “left against medical advice” (AMA), which is a distinct disposition that happens after a patient has been seen. Boarding and LWBS are causally linked: when boarded admitted patients occupy treatment spaces and hallway capacity, incoming patients wait longer for a bed and a bigger share of them leave before being seen — which is why the two metrics are reported together rather than in isolation.

3. Door-to-provider time (time to diagnostic evaluation)

The interval from ED arrival to first evaluation by a qualified medical professional (physician, PA, or NP). This one measures the front end of the ED visit rather than the back end, but it belongs on the same dashboard as boarding because the two interact: a department with severe exit block (patients unable to leave the ED for an inpatient bed) backs up treatment space, which pushes door-to-provider time out for everyone still arriving. A hospital that only watches boarding time and ignores door-to-provider can miss the earliest signal that boarding is starting to degrade the rest of ED throughput.

4. Median ED arrival-to-departure time, split by disposition

Total ED length of stay, reported separately for admitted versus discharged patients, because the two populations have very different drivers. Discharged-patient length of stay reflects workup and treatment efficiency; admitted-patient length of stay is dominated by boarding once the admit decision is made. Reporting a single blended ED length-of-stay number obscures which problem a hospital actually has — this split is what makes the metric actionable rather than descriptive.

How These Metrics Tie to CMS ED Throughput Measures

These aren’t metrics CASRAI or individual hospitals invented independently — they trace directly to a measure set CMS built for the Hospital Outpatient Quality Reporting (OQR) Program specifically to standardize ED throughput reporting nationally:

• OP-18b — median time from ED arrival to ED departure for discharged ED patients (the discharged-side counterpart to the metric above).
• OP-20 — door to diagnostic evaluation by a qualified medical professional (the same definition as door-to-provider time above).
• OP-21 — median time to pain management for patients with a long-bone fracture.
• OP-22 — patient left without being seen, before being evaluated by a qualified medical professional (the LWBS measure above, in CMS’s exact specification).

CMS periodically removes measures from the OQR Program’s mandatory public-reporting set once they’re judged “topped-out” or otherwise no longer meet the program’s selection criteria, and the ED throughput measures have been affected by that process at various points — so don’t assume all four are currently required for federal reporting without checking the current-year Hospital OQR specifications manual on QualityNet before citing one as mandatory. What hasn’t changed is that these definitions remain the standard vocabulary for ED throughput measurement: state hospital associations, ACEP’s benchmarking work, and peer-comparison collaboratives all use the same numerator/denominator logic CMS established, which is exactly why building your internal dashboard on these definitions — rather than an ad hoc local variant — keeps your numbers comparable to peers even as a given measure’s federal reporting status changes.

The Joint Commission Patient Flow Standard

Boarding measurement isn’t only a CMS reporting question — it’s a Joint Commission leadership standard. LD.04.03.11 requires hospital leadership to develop and implement plans to manage the flow of patients throughout the organization, and it names ED boarding specifically as part of that obligation. Following its 2022 Sentinel Event Alert on the safety risks of ED boarding — which documented the mortality, delayed-treatment, and privacy harms associated with extended boarding, with particular emphasis on behavioral health patients, who tend to board longest and are least well served by hallway-bed care — the Joint Commission built an expectation into the standard that when average boarding time crosses a defined threshold, hospital leadership must be actively involved in identifying and implementing alternatives, not simply monitoring the number. Because accreditation standards and their elements of performance are periodically revised, verify the current text of LD.04.03.11 against the Joint Commission’s current E-dition before citing a specific hour threshold or EP number in survey-preparation materials — the obligation itself (leadership engagement tied to a boarding-time trigger) is the stable part; exact wording is what to double-check.

The practical implication for a quality program: your boarding-time dashboard isn’t just an internal improvement tool. It’s the evidence base a surveyor will expect to see when assessing whether leadership is actually managing patient flow under LD.04.03.11, which is a second, independent reason (beyond the CMS measure history above) to keep the underlying data collection rigorous and auditable rather than an informal spreadsheet a charge nurse updates by hand.

Building a Boarding Dashboard That Holds Up to Audit

Three data-quality issues account for most of the boarding-metric disputes a quality team runs into:

• Timestamp source. Admit-decision time and bed-arrival time should come from the same system of record (typically the ADT feed inside the EHR), not a mix of EHR timestamps and a manually logged bed-board time — the two frequently disagree by 20-40 minutes, and using both inconsistently across cases makes trend data unreliable.
• Denominator scope. Decide up front whether observation patients, psychiatric holds awaiting an inpatient behavioral-health bed, and inter-facility transfers count in the boarding denominator, and apply that rule consistently — these are exactly the populations most likely to be quietly excluded (intentionally or not) in a way that flatters the topline number.
• Outlier handling. Report median and 90th-percentile together (see above), and flag — rather than exclude — cases boarding past a defined ceiling (commonly 12 or 24 hours) for individual root-cause review. Excluding them from the aggregate metric because they’re “unusual” is how a dashboard stops reflecting the patients most at risk.

Interventions Hospitals Use to Reduce Boarding

ACEP’s boarding and crowding resources and IHI’s patient-flow collaborative work both document a consistent set of interventions, though the reported effect sizes vary substantially by hospital size, case mix, and baseline severity — treat any single published percentage as directional rather than a number your hospital should expect to replicate exactly:

• Full-capacity protocols — moving boarded, stable admitted patients to hallway spaces on inpatient units instead of the ED, on the documented logic that hallway boarding is safer and less resource-intensive when distributed across a whole hospital’s capacity than when concentrated in one department.
• Bed-management/patient-flow coordinator roles — a dedicated real-time function (sometimes called a “bed czar”) whose job is actively matching admitted patients to open beds rather than leaving placement to unit-level negotiation, which is the step that tends to add avoidable delay during high-census periods.
• Surgical and discharge-timing interventions — smoothing elective surgical scheduling and moving discharge decisions earlier in the day, both aimed at the same root cause: bed availability is a whole-hospital flow problem, and ED boarding is usually the most visible symptom of it rather than an ED-specific failure.
• Behavioral-health-specific pathways — dedicated psychiatric assessment and disposition processes, given that behavioral-health patients consistently board longest and are the population the Joint Commission’s alert specifically flagged.

A durable reduction in boarding time almost always requires action outside the ED — inpatient discharge timing and bed-turnover speed are usually bigger levers than anything an ED can control on its own, which is why boarding metrics belong on a hospital-wide flow dashboard, not just an ED operations report.

Frequently Asked Questions

What is considered “boarding” in an emergency department?

A patient is boarding once an admit decision has been documented but they remain physically in the ED because no inpatient bed, OR slot, or transfer is yet available. The clock starts at admit-decision time, not at ED arrival.

Is the left-without-being-seen rate the same as patients leaving against medical advice?

No. LWBS refers to patients who leave before ever being evaluated by a provider; leaving against medical advice (AMA) happens after a patient has been seen and is a distinct disposition category. Track them separately.

Does CMS still require hospitals to report ED throughput measures like OP-18 and OP-22?

The specific set of ED throughput measures required under the Hospital OQR Program has changed over time as CMS retires “topped-out” measures. Confirm current requirements against the live Hospital OQR specifications manual on QualityNet rather than assuming a given measure’s past status still applies — but the underlying definitions remain the field’s standard vocabulary regardless of current federal reporting status.

What does the Joint Commission’s patient flow standard actually require?

LD.04.03.11 requires hospital leadership to manage patient flow throughout the organization, including ED boarding, and — following the Joint Commission’s 2022 Sentinel Event Alert on boarding risk — ties sustained boarding above a defined threshold to a leadership-engagement expectation, with particular attention to behavioral-health patients. Verify the current standard text and any specific hour threshold against the Joint Commission’s current E-dition before using it in survey preparation.