Fixing the emergency department: Why knowing what to do isn’t enough 

The operational practices that improve ED throughput are well established. So why do so many improvement efforts stall? After years working alongside hospital leaders, the pattern is clear — and so is the path forward.

The median wait time in an American emergency department is 168 minutes. [1] That alone is a difficult number to sit with. But the more revealing statistic is the spread around it: approximately 15% of hospitals report median visit times exceeding four hours, and around 4.5% exceed five. [2] Entire geographies report systematically slower performance: median ED visit times in Washington D.C., Maryland, Massachusetts, and Rhode Island run from nearly an hour to more than two hours longer than the national average. [3] The median rate of patients leaving without being seen is 2%, yet about 15% of hospitals run at double that rate or higher. [4] 

The gap between average performers and struggling ones is not explained by who has access to better operational practices. The practices are largely the same, widely published, and broadly understood. The gap is explained by who can implement them, sustain them, and continuously refine them — and who cannot. 

The pressures are real, and they are compounding 

Emergency departments are operating under conditions that would have been hard to imagine a decade ago. Patient volumes continue to rise. Workforce shortages have not meaningfully eased. Clinical complexity is increasing as patients present later and sicker. And inpatient capacity, the release valve for ED flow, remains tight at most organizations. 

The downstream effects are familiar to anyone who has walked through a busy ED at 7 p.m.: hallway boarding, frustrated clinicians, families waiting for hours, and quality metrics that drift in the wrong direction. These pressures do not stay contained in the ED. They radiate outward, affecting patient experience scores, workforce engagement, and financial performance across the organization. 

The misconception: That we don’t know what works 

Over the past decade, hospitals have generated a substantial body of evidence about what improves ED performance. Fast-track lanes for low-acuity patients. Provider-in-triage models. Disciplined diagnostic turnaround standards. Inpatient pull systems. Early discharge planning. Hospitalist alignment. Surge protocols that activate before the ED is already underwater. 

These practices are not secrets. Most ED medical directors and patient flow executives can list them from memory. Yet performance variation persists — often dramatically — between hospitals that, on paper, are pursuing the same playbook. 

The challenge is not knowing what to change. The challenge is making the change actually take hold. 

Where improvement efforts actually break down 

In the work we do alongside hospital leaders, four patterns explain most of the failed or stalled initiatives we encounter. 

1. The wrong solution for the environment 

A split-flow, direct-to-room model can dramatically accelerate throughput at facilities with the right provider coverage and physical layout and create congestion and confusion at hospitals where the footprint or staffing model cannot support it. Selecting the right combination of operational changes requires a thoughtful read of patient volumes, staffing models, physical capacity, organizational culture, and the broader hospital operating model. Skip that diagnostic step, and even well-regarded practices can underperform. 

2. Adopting the form without the function 

Closely related, and perhaps more common, is the failure to translate the spirit of a best practice into actual operational design. A patient flow executive is named but never granted the authority required to direct movement across departments. A fast-track lane is built, but the provider rotation and staffing model around it is not adjusted to accelerate throughput. A hospitalist “quarterback” role is established, but its accountabilities and incentives are not designed to genuinely speed admissions and movement to the floors. The structures exist on paper. The performance they were designed to produce does not. Best practice becomes a label rather than a working mechanism. 

3. Adoption without ownership 

New workflows often get introduced through a memo, a training, and a go-live date. What is missing is the work of helping frontline teams understand why the change is happening, how it will affect their daily responsibilities, and what role they play in the outcome. Without that grounding, adoption becomes inconsistent — some shifts run the new process, others quietly default to the old one — and the improvement never fully materializes. This pattern also tends to amplify the previous one. When adoption is treated as a check-the-box exercise rather than a genuine behavior change, teams execute the surface mechanics of the new workflow without absorbing its purpose, and the outcomes the work was designed to produce remain out of reach. 

4. The slow drift after go-live 

Even well-designed processes drift. Staff turnover. Volumes shift. Workarounds creep in. Without a structured way to monitor whether the change is still producing the intended results, leaders often discover months later that the gains they reported at go-live have quietly eroded. Implementation is treated as the finish line when it is actually the starting line. 

What sustainable change actually requires 

Organizations that achieve durable improvement in ED performance tend to share a set of disciplines. None of them are exotic. All of them require deliberate work. 

Inclusive solution design. The people who will execute the new workflow — nurses, physicians, technicians, support staff, ED leadership — are involved in shaping it. This is not consensus theater. It is the most reliable way to surface operational realities early and build the ownership that adoption requires. 

Structured change management. Communication, stakeholder engagement, education, and deliberate training are planned with the same rigor as the workflow redesign. Education ensures teams understand what is changing and why it matters. Training and competency development ensure they have the skills to execute the new work consistently. Both are required — knowing is not the same as doing. 

At-the-elbow implementation support. During go-live and the weeks that follow, someone is physically present to help operational and physician leaders work through barriers in real time. This is where most of the small problems that would otherwise compound get caught and corrected. 

Continuous performance monitoring. Defined metrics, regular review cadences, and a feedback loop that converts data into adjustments. The goal is to hardwire the new process into daily operations until the desired behaviors occur consistently — then keep watching to make sure they continue to. 

The breadth of the opportunity 

When we work with hospitals on ED and patient flow improvement, the work typically spans six interconnected areas: getting patients into the right care pathway from the moment they arrive; moving them efficiently through the ED with optimized provider workflows and diagnostic turnaround; creating inpatient capacity through structured pull models and discharge readiness; coordinating care and accelerating disposition for observation and complex patients; streamlining admissions, handoffs, and transfers between ED and inpatient teams; and responding effectively during periods of high demand through forecasting, surge protocols, and enterprise-wide accountability. 

Few organizations need to work on all six at once — and most should not try to. The diagnostic work upfront is what identifies which areas will produce the most leverage for a given hospital. In the engagements we have led, a recognizable pattern emerges hospitals tend to capture the majority of their gains from a focused set of two or three well-scoped initiatives matched to their environment, rather than from working broadly across all six. 

Why it is worth the effort  

The organizations that get this right realize benefits that extend well beyond the ED itself. Faster access to care and reduced LWBS rates captures revenue that was previously walking out the door. Reduced boarding and improved throughput for time-sensitive conditions improve clinical quality. Frontline teams report less frustration and stronger collaboration between ED and inpatient services. Patient satisfaction scores recover. The financial performance of the ED improves — and so does the performance of the hospital around it. 

There is a virtuous cycle available to organizations that commit to this work: patients receive care more quickly, staff are better supported, quality outcomes improve, and operational and financial performance follow. The practices that produce that cycle are known. The work is in choosing the right ones, implementing them with discipline, and refusing to treat go-live as the finish line. 

Sources 

[1] CMS Timely & Effective Care – Hospital dataset, measure OP_18a (median time all patients spent in the emergency department, including psychiatric/mental health and transfer patients), reporting period April 2024 – March 2025.

[2] Calculated from the hospital-level CMS Timely & Effective Care dataset described in note 1. 

[3] CMS Timely & Effective Care – State dataset, calendar year 2024 (measure OP_18b, which excludes psychiatric/mental health and transfer patients; OP_18a state-level breakdowns are not separately published by CMS). 

[4] CMS Timely & Effective Care – Hospital dataset, measure OP_22 (left without being seen), self-reported by hospitals via CMS’s Hospital Quality Reporting (HQR) Secure Portal, hospital fiscal year range January 1, 2024 – March 31, 2025; data retrieved March 13, 2025.