To address the patient safety risks associated with provider handoffs, Seattle Children’s Hospital designed a new staffing model that is built around waterfall-style shifts, in which a new attending physician arrives every three to five hours. The approach has significantly reduced the number of handoffs that occur without increasing attending physician hours. Further, it has won the approval of both providers and charge nurses.

  • Under the new staffing model, provider shifts are staggered so that handoffs do not all occur at the same time, relieving some of the end-of-shift burdens on physicians at a time when they are likely tired.
  • The model frontloads most of a provider’s work in the first few hours of a shift, tapering these responsibilities as new providers arrive. This reduces the number of handoffs required at the end of a shift, particularly handoffs involving complex patients.
  • In a retrospective analysis of nearly 44,000 patient encounters that occurred under the new staffing model, researchers found that there was a 25% reduction in encounters that included patient handoffs from one provider to another.
  • Experts note the approach can be scaled to work in an ED of any size that provides more than single-provider coverage.

Patient handoffs are a point of risk for medical errors in the emergency setting. For example, during shift changes, an outgoing provider may be stressed and fatigued, heightening the potential that he or she may fail to highlight important information or miscommunicate clinical details about one or more patients. Further, errors of this sort can be compounded if incorrect information is documented and continues to follow a patient as he or she moves through the health system.

Recognizing the risks to patient safety, many investigators have focused on improving handoff processes to reduce the potential for errors when the care of patients is transferred to new providers. This is valuable work, but what if someone takes steps to address the number of handoffs that occur as well as the stress and fatigue that clinicians often experience toward the end of their shifts?

Investigators at Seattle Children’s Hospital have found that by adjusting their staffing model, they can not only reduce the number of handoffs that take place in the ED without increasing attending provider hours, they also boost provider satisfaction. Further, while the staffing model takes work and patience to implement fully, researchers believe that with the proper support and commitment, the approach is flexible enough to work in many other EDs, both adult and pediatric.

The opportunity to make a change around the scheduling of providers presented itself to Seattle Children’s leadership when the ED transitioned into new quarters in 2013, explains Hiromi Yoshida, MD, MBA, an emergency attending physician there and a clinical assistant professor of pediatric emergency medicine at the University of Washington School of Medicine. Yoshida was the lead author of an investigation into the development, deployment, and results of the new staffing model.1

“We wanted to improve patient care, but at the same time we also wanted to make the experience better for the physicians that work here,” she says.

Under the original staffing model, the ED consisted of three pods with seven shifts that each lasted seven to nine hours. “The idea was that at the end of your shift, you signed out to the next person who was coming in. Basically, you had to handoff everyone that you had to the next person ... when your shift ended,” Yoshida explains.

It is a fairly traditional approach with which many emergency providers likely are familiar. However, this approach places considerable demands on clinicians at the very end of their shifts when they tend to be fatigued. It is at this point when they need to communicate important information about their patients to the next provider who will be taking over their care. In the ED at Seattle Children’s, there were several concerns about this original staffing approach, Yoshida shares.

“Patient care just kind of stopped for a little bit of time [during the handoffs]. There were delays because people had to wait, and I think that led to frustration from the staff ... patient care wasn’t moving forward. Potentially, that could lead to patient harm,” she says. “Also, people were working at full capacity throughout their entire shift. We know that people make better decisions toward the beginning of their shifts because decision-fatigue kicks in toward the end.”

The new staffing approach, designed by a multidisciplinary team, attempts to ease some of these end-of-shift burdens by collapsing the model into two pods and staggering the arrival of oncoming attending physicians in what creators of the approach describe as waterfall shifts. Under this structure, when a physician first arrives for work, he or she is considered the primary attending provider and begins seeing patients.

“The next attending physician arrives three to five hours later, depending on the shift. Then, [he or she] becomes the primary attending,” Yoshida explains.

At this point, the provider who arrived first becomes the secondary attending. In this secondary attending role, Yoshida says this physician can complete work on existing patients. Meanwhile, secondary attendings can work with other physicians on less complex cases, treating and discharging those patients before the end of the secondary attendings’ shifts.

The thinking behind this approach is that by the time a provider’s shift ends, there will at least be a plan in place for any patients still under the clinician’s care. These patients are likely less complex because of the way the provider picked up patients. Also, there are likely to be fewer handoffs, Yoshida notes.

“For those patients who do need to be handed off, there will be fewer opportunities for miscommunication because [the patients] will hopefully be less complex, less sick,” she says.

Further, with a new provider coming on board every few hours, leaders hoped that the new waterfall-style shifts would provide relief and a morale booster to emergency clinicians, Yoshida says.

“When providers come in, they can just start seeing patients. No one has to wait to staff their patients. Everything just keeps moving forward,” she says. “Sign-out happens at staggered times. It is not necessarily happening with the whole board [of patients]. You are signing out fewer patients, which is less stressful, I think, for the people doing it.”

Following implementation of the new staffing model in April 2013, a retrospective study by Yoshida and colleagues found that the overlapping, waterfall-shift approach significantly reduced handoffs compared to the previous staffing model. In an analysis of nearly 44,000 patient encounters that occurred under the new staffing model, there was a 25% reduction in encounters that included patient handoffs from one provider to another.

Researchers also observed improvements on secondary metrics. For example, while there was a very slight increase in average length of stay under the new staffing model, the left-without-being-seen rate and the 72-hour return rate for patients both decreased slightly. Also, the percentage of charts signed within 72 hours increased from 90.4% to 95.7%.

Investigators noted that there was one serious safety event both before the change to the new staffing model and following implementation. However, when surveyed about the new model, physicians and charge nurses reported that, in their view, the new model improved patient safety. Twelve of 18 attending physicians surveyed reported that they “strongly agreed” that the model improved patient safety, and six attending physicians responded that they “agreed” with the statement.

Similarly, among 13 charge nurses who responded to the survey, eight indicated they “strongly agreed” that the model improved patient safety, and five charge nurses indicated they “agreed” with the statement. The surveyed charge nurses also indicated that the new model improved patient flow in the ED.

When given an opportunity to provide comments about the model on the survey, some physicians voiced strong approval of the waterfall-staffing approach.

“One physician commented that the staffing model is the single most important change we made in terms of contributing to job satisfaction. Someone else mentioned that it had definitely increased their longevity [in the job],” Yoshida reports. “I do think the way it is set up will keep experienced physicians here longer and also keep them happier ... having happy, healthy staff leads to better care.”

While the waterfall shifts are firmly ingrained in the ED at Seattle Children’s, the clinicians who championed the model note that others interested in adopting the method should not underestimate the challenges involved with implementing a new approach.

“The biggest challenge for us was that people are generally resistant to change, so we did a lot of talking to individuals and groups before the change to get their buy-in,” explains Susan Mazor, MD, an emergency attending physician at Seattle Children’s, an assistant professor in pediatric emergency medicine at the University of Washington School of Medicine, and one of Yoshida’s colleagues on the study of the waterfall handoff structure.

Yoshida echoes these sentiments, noting that it is important to collect feedback from everyone about any proposal. “However, remember that sometimes you just have to try it and see how it goes in the spirit of quality improvement,” she says. “Emergency medicine is a great place given that everyone is flexible and willing to pitch in ... for quality improvement work, the ED is a very willing place to do things like this.”

The new staffing approach should be able to work in adult and pediatric EDs, but there does need to be enough staff to construct the waterfall shifts.

“The plan should be able to be scaled to an ED of any size with more than single coverage,” Mazor offers.

As with any staffing model, changes in volume and other factors need to be addressed. The ED at Seattle Children’s has continued to tweak its approach.

“Last winter was one of the busiest winters we have ever had. As our census grows, we are actually adding a shift during the winter months so we can see more patients,” Yoshida explains, noting that adding more staff is not only for the health of patients but also to maintain the health of ED staff.

Further, while the model has been very well received, ED leadership continues to collect feedback and to make improvements as needed, although the approach has proven to be robust and flexible.

“There are more physicians during our peak hours. The waterfall starts with only one physician. As the day goes on, more physicians come more frequently ... we try to match the timing based on when we think patients are coming,” Yoshida notes. “Ninety percent of the time, the model works well because of the way the waterfall shifts are scheduled. We have the [highest] number of physicians when we know we are going to be the busiest.”


  1. Yoshida H, Rutman LE, Chen J, et al. Waterfalls and handoffs: A novel physician staffing model to decrease handoffs in a pediatric emergency department. Ann Emerg Med 2018; Oct 1. pii: S0196-0644(18)31160-0. doi: 10.1016/j.annemergmed.2018.08.424. [Epub ahead of print].


  • Susan Mazor, MD, Emergency Attending Physician, Seattle Children’s Hospital; Assistant Professor, Pediatric Emergency Medicine, University of Washington School of Medicine. Email: susan.mazor@seattlechildrens.org.
  • Hiromi Yoshida, MD, MBA, Attending Emergency Physician, Seattle Children’s Hospital; Clinical Assistant Professor, Pediatric Emergency Medicine, University of Washington School of Medicine. Email: hiromi.yoshida@seattlechildrens.org.