Why the highest-leverage ED throughput solutions are also the hardest to implement

The practices that improve ED throughput are well known. What separates the hospitals that succeed is willingness to do the structural and cultural work each high-impact intervention requires.

Two community hospitals of comparable size and acuity mix can post emergency department wait times that differ by hours. Two academic medical centers in the same metropolitan area can run left-without-being-seen rates that differ by orders of magnitude. The gap between strong and struggling ED performers is rarely explained by which institutions have access to better operational practices. The practices that improve ED flow are well established, widely published, and broadly understood. Performance varies because implementation does. 

In our work with hospitals on ED throughput, a small number of interventions consistently produce disproportionate impact when they are implemented well. They are the solutions most worth investing in. They are also, in most cases, the solutions that face the most cultural and structural resistance during implementation — which is precisely why so many organizations either avoid them or adopt them in name only. 

Six interventions, in particular, do most of the work. 

1. Getting patients into the right care pathway 

Few interventions move ED performance as reliably as the right routing decisions at the front of the department. Three connected front-end strategies drive most of the throughput benefit, and each carries its own cultural complexity. 

Fast-track lanes for low-acuity patients 

A fast-track lane peels off low-acuity patients — typically ESI 4 and 5 — into a dedicated workflow with staffing matched to that acuity. In most contemporary models, that means an investment in advanced practice provider (APP) coverage: a nurse practitioner or physician assistant supported by an emergency physician. APP coverage has become common in U.S. emergency departments, present in roughly two-thirds of EDs by recent counts [1], but the implementation quality varies widely — and so does the resistance to making the investment work as designed. 

Several tensions come up consistently. Physician groups can have legitimate concerns about supervision standards and scope-of-practice clarity, particularly where state regulation leaves room for interpretation. Some research suggests that APP-staffed EDs show higher rates of imaging utilization and admissions than physician-only departments, complicating the cost case for the investment [2]. Some organizations end up assigning APPs to below-license tasks — paperwork, charting support, low-complexity follow-up — that erode the throughput benefit. And in some hospitals, the physician group views the model as a productivity threat rather than a collaborative partner. 

The hospitals that get fast track right have generally done the hard work of defining the APP role explicitly: clear protocols for which patients route to fast track, clear supervision standards, clear scope expectations, and a compensation and productivity model that rewards collaboration rather than competing for credit. 

Vertical flow and results-pending areas 

For higher-acuity patients awaiting workup — labs, imaging, specialty consultation — the conventional model has each patient occupying a stretcher in a treatment room from arrival to discharge. A vertical flow model substitutes recliners or stretcher-chairs in a designated results-pending area, freeing the treatment room for the next patient as soon as initial evaluation, orders, and any necessary procedures are complete. Reported length-of-stay reductions range from modest to substantial — 25% or more in well-designed implementations [3]. 

Resistance to vertical flow is more often perceptual than operational. Patients and families may feel less cared for sitting in a recliner than lying in a stretcher, even when the clinical care delivered is identical. Staff and physicians sometimes share that discomfort. Reimbursement concerns surface periodically, though when the model is designed properly it accommodates standard ED billing requirements. 

Successful implementations address the perception issue directly: designing the results-pending area to feel like clinical space rather than a waiting room, ensuring visible nursing presence, providing privacy screens, and communicating clearly to patients about why they are there and what comes next. 

Split-flow and direct-to-room 

A split-flow model routes higher-acuity patients directly from triage to an available bed, bypassing the secondary waiting room, while lower-acuity patients route to fast track or vertical flow. The model typically requires a provider-in-triage — often called a “doctor in triage” — to make the routing decision: which patients are acute enough to go straight to a bed, which can be safely worked up vertically, and which belong in fast track. 

Cultural resistance to this design comes from two directions. The provider-in-triage role itself is unpopular with some emergency physicians, who view it as less satisfying than traditional bedside care and worry that triage-pace decisions may not reflect their best clinical judgment. Separately, staff and physicians sometimes feel uncomfortable taking patients directly from triage to an open bay without the customary rooming workflow — vital signs reconciliation, history confirmation, initial workup orders — being completed first. The workflow change feels rushed, even when the clinical evidence supports it. 

Hospitals that succeed with split-flow have generally invested in physician group buy-in early, established clear protocols for what counts as direct-to-room and what does not, and redesigned the rooming workflow rather than simply asking staff to skip it. 

2. Empowering the patient flow executive 

Many hospitals have created this role. Fewer have given it the authority required to be effective. A patient flow executive who can only suggest, escalate, and hope is structurally limited. The role needs the standing to ensure designed processes are followed across departments — and the authority to move patients consistent with the operating model when they are not. 

The cultural complexity is rarely about whether the role exists. It is about whether senior leadership has empowered the role to make decisions that override the natural inclinations of individual departments. Without that empowerment, the role becomes a coordinator of dysfunction rather than a driver of flow. 

Organizations that get this right tend to share a few characteristics: visible executive backing for the role, defined decision authority around bed assignment and patient placement during high-census conditions, and a direct reporting line to senior leadership rather than burying the position inside operations. 

3. Timely disposition of observation patients 

Observation patients are an underappreciated lever in ED throughput. They are medically stable enough to leave the ED but require a defined period of monitoring before final disposition. The question is where they go and how their status is resolved. 

The common approach — pushing observation patients to inpatient floors as space allows — feels efficient. In practice, observation patients can get absorbed into the rhythm of medical-surgical units where they do not fit the typical patient profile or rounding cadence. The intended 12-to-19-hour observation window stretches to 24 hours or more. Beds remain occupied. The throughput benefit is partially or fully offset as bed capacity is held for far longer than the patient’s clinical needs require. 

The financial implications go beyond throughput. An efficient observation process — one that resolves observation status within target time windows — can also reduce avoidable inpatient admissions, generating financial benefit on top of the operational improvement. 

The more effective approach varies by hospital but usually involves some combination of a dedicated observation unit with defined staffing and protocols, clear ownership for resolving observation status (typically a hospitalist or designated provider with explicit accountability for disposition within target time windows), visible metrics tracking observation length-of-stay, and discharge processes designed for the observation pace rather than the inpatient pace. 

The cultural complexity here is real. Observation patients sit between two operating models, and most hospitals have not picked one. Hospitalists may resist clear ownership; nursing units may resist dedicated observation expectations; finance teams may be ambivalent about the operating design that supports the observation revenue model. The result is patients who linger because no one is structurally accountable for their movement. 

4. Direct ED–hospitalist alignment 

Organizations frequently steer clear of the work that must happen directly between emergency physicians and the hospitalist team. The work is uncomfortable — it touches admission criteria, communication standards, accountability, and sometimes compensation models — but smooth, timely handoffs cannot be wished into existence. They require an explicit operating model that both groups have built together and committed to. 

Two specific implementations tend to drive outsized gains, and both tend to draw outsized resistance. 

Bridge orders 

When hospitalists are willing to write bridge orders — covering essential care needs during the gap between admission decision and inpatient bed assignment — patients move out of the ED faster, even when they are temporarily boarded. Resistance to bridge orders is typically rooted in workflow concerns and scope-of-practice ambiguity. Both are solvable with explicit operating agreements between the ED and hospitalist groups, but the conversation has to happen. 

Co-location of a hospitalist quarterback in the ED 

A hospitalist physically located in or near the ED — designated to evaluate admissions, write orders, and coordinate disposition in real time — can transform handoff timeliness. The model is not appropriate for every hospital; it requires admission volume to justify the staffing investment. But where the volume supports it, the impact is substantial. Resistance typically centers on hospitalist group concerns about workflow disruption and the perceived loss of autonomy that comes with being embedded in another department’s operational environment. 

The common thread across both is that ED–hospitalist alignment requires explicit, intentional design between two groups that often prefer to operate independently. Avoiding this conversation is one of the most common reasons handoff delays persist year after year. 

5. Hospitalist and nursing unit ownership of pull 

A pull system only functions when the receiving units actively pull. That means the hospitalist team and the inpatient nursing units share ownership of moving admitted patients out of the ED and onto the floors in a timely way. 

Alignment tools to create that ownership exist — visible flow metrics, dyad accountability between nursing and hospitalist leadership, structured expectations around bed assignment and patient acceptance, thoughtful incentive design — but they are seldom implemented with the rigor the work requires. 

The cultural complexity is the work itself. Pull systems require receiving units to absorb work they would naturally prefer to defer during periods of high demand. The institutions that build effective pull systems have made the receiving units’ performance on patient acceptance and bed turnover a visible, measured, and accountable element of operations — with consequences attached. 

6. Shared ownership of surge response 

Surge escalation protocols work when inpatient leaders, nursing units, and physician teams view them as a hospital-wide responsibility for protecting patient access rather than as an ED concern that periodically asks the rest of the organization for help. 

That shift is partly structural and partly cultural. It is built by leadership that consistently treats ED capacity as a measure of how well the whole institution is functioning, and by escalation protocols that name specific roles and actions outside the ED rather than leaving the inpatient response to discretion. 

Well-designed surge protocols also formally extend the authority of the patient flow executive at escalated levels. At an orange or red surge status, the patient flow executive can be empowered to make decisions about bed assignments, discharges, and unit-level patient acceptance that would otherwise be subject to negotiation. These protocols typically also engage the Chief Medical Officer or a clinical equivalent as part of the escalation, ensuring that the most consequential surge decisions are made with clinical leadership backing rather than as operational requests. 

Done well, it is transformative. Done as a memo, it is ignored. 

Where to Start 

Not every hospital needs to pursue all six interventions, and most should not try to. The diagnostic work upfront identifies which two or three offer the most leverage given current patient volumes, physical capacity, organizational culture, and the relationships already in place between physician groups and operating leadership. 

What unites the six is implementation difficulty, not knowledge gaps. Each has been studied, written about, and successfully implemented at peer hospitals. What separates organizations that achieve durable gains from those that do not is not whether they recognize these interventions but whether they are willing to do the structural and cultural work to make them function as designed. 

That work — specific, uncomfortable, and rarely glamorous — is where the throughput gains live. 

Sources 

[1] Emergency Department Benchmarking Alliance (EDBA) data, as reported in Aledhaim A, Walker A, Vesselinov R, Hirshon JM, Pimentel L. “Resource Utilization in Non-Academic Emergency Departments with Advanced Practice Providers.” Western Journal of Emergency Medicine. 2019;20(4):541-548.

[2] Aledhaim A, Walker A, Vesselinov R, Hirshon JM, Pimentel L. “Resource Utilization in Non-Academic Emergency Departments with Advanced Practice Providers.” Western Journal of Emergency Medicine. 2019;20(4):541-548. Study found APP-staffed EDs were associated with 30.4 more admissions per 1,000 visits and modest increases in imaging utilization compared to non-APP-staffed EDs.

[3] American College of Emergency Physicians (ACEP), Quality Improvement & Patient Safety Section newsletter, “Optimizing Throughput in the Emergency Department: One Institution’s Experience,” Winter 2025.