Emergency Departments

In 2002, Mary Washington Hospital in Fredericksburg, Virginia, thought it had solved all of its emergency department (ED) problems by doubling its size to 50 beds and 26,000 state-of-the-art square feet. It had seen 75,000 patients that year, with a 6% walkout rate. A year later, patient volume had climbed to 83,000, but all beds were in constant use and the walkout rate had climbed to 14%, including 3,000 walkouts over a four-month period.

Patients arriving with no more than earaches or sore throats—the primary care cohort that most EDs commonly see these days—were waiting more than 2.5 hours for attention. Three years later, a young ED physician, Jody Crane, despairing he could never be maximally effective in such a setting, began studying for an MBA and a possible career in business. He learned of Toyota Motor Corp.’s Lean manufacturing processes—and the rest is ED history.

By 2009, the walkout rate at Mary Washington had dropped to below 2% despite an increase in volume to 100,000, while length of stay declined from four to fewer than three hours. The ED’s average wait time was 25 minutes.

Crane’s Lean work (more later), which has since included consultations with hospitals throughout the United States and authorship of the book “The Definitive Guide to Emergency Department Operational Improvement,” is only one harbinger of a new era in ED design. With ED volumes increasing steadily by some 2% a year over the past 20 years, and a growing concern that volumes might escalate geometrically with the advent of healthcare reform, along with the reasonable notion that the ED, generating more than 50% of admissions, is in fact the hospital’s “front door,” the ED has become a focus of healthcare design innovation.

Frank Zilm, principal of planning and design firm Frank Zilm & Associates and a leading advocate of ED redesign, says, “The ED is going through some of the most exciting operational transitions in healthcare. We’re eliminating waste, streamlining processes, managing low-acuity patients more efficiently than ever, abandoning triage for team assessment, and, in general, doing a complete rework of the front end of the hospital. The concepts emerging from this will be accepted by most EDs within the next five years.”

The principal goals of sorting patients into appropriate areas of the ED and getting each cohort as close as possible to needed resources are guiding these efforts. For example, the traditional “ballroom” layout so familiar to many, with examination rooms surrounding a common work/observation area, seems to have hit its useful limits with a maximum of 16-18 exam rooms.

Beyond that, the optimum visibility the layout promotes fades, and its basic inflexibility in terms of patient volumes and staff initiative becomes obvious. A more recent alternative has been the “pod” layout, dividing the ED into clinically specialized units but still facing difficulty “flexing” up or down with patient population, and the linear model, with parallel units bordering a central work area; this allows maximum flexibility but requires a commitment to decentralized staff.

More recent variations on these models have emerged, including James Lennon’s “matrix” layout, interspersing exam rooms with work areas for more efficient adjacencies and—as we’ll see in a moment—universally equipped “care zones.”

The Mary Washington Hospital experience, in added detail, helps show how all of this came about and where it is heading. Crane’s redesign team began in 2006 by focusing on the lowest-acuity patients entering the ED—the earaches, the sore throats, and so forth. The team created a “Super Track” with two or three beds and a small dedicated team of nurses and technicians examining and treating patients within an hour. This required relocating equipment storage so that the types and numbers of medical equipment needed for this particular group were readily available only a few steps away. Central storage eventually came to be a thing of the past. From this change alone, the length-of-stay numbers (LOS) and improved patient flow were dramatic.