“Mr. Overton? Mr. Overton?!” Nurse Rusty DeGuzman’s patient in the intensive care unit at Stanford Hospital & Clinics was vomiting blood and responding erratically at 10:01 p.m. DeGuzman ordered six units of blood, and one minute later pulmonary critical-care fellow Doan Luu, MD, arrived. Luu ordered a “massive transfusion protocol” from the blood bank, and added, “We need to intubate.”

Within minutes, an anesthesiologist came in with the anesthesia airway box, as Luu prepared to put in a central line, or IV, to deliver medication. By 10:12 p.m. a respiratory therapist was hand-bagging Mr. Overton, and eight other professionals were gathered around his bed, monitoring his pulse, blood pressure and breathing.

“Okay, that’s it,” Geoff Lighthall, MD, PhD, announced at 10:14 p.m. To his colleagues running the event with him, he remarked, “That was a high-performing crew.”

On a typically busy night, ICU nurses and physicians had interrupted whatever they were doing to care for Mr. Overton. They knew right away that he wasn’t a typical patient—his plastic torso was a giveaway—but no one cracked a grin or whispered the words “mock” or “mannequin.” As far as they were concerned, it was the real deal.

In the pilot year of a program designed to test Stanford Hospital’s response to critical, life-threatening events, the recent mobile simulation exercise was the seventh of 12 planned exercises. Directed by Lighthall, associate professor of anesthesia at the School of Medicine, the unannounced drills are designed to “stress test” the hospital’s emergency response systems, according to Jeff Driver, the hospital’s chief risk management officer.

“Getting blood to a patient’s bedside sounds so simple, but there’s actually a series of steps that must take place, and at any point things can go wrong,” Driver said. “So we stress our system to understand where the vulnerabilities are, to expose them and clean them up. The idea is to allow ourselves to make errors in a lab environment, so that we’re not making them when we’re caring for patients.”

Driver and Lighthall will present initial findings from Stanford’s simulation exercises on Oct. 3 at the annual conference of the American Society for Healthcare Risk Management meeting in Boston. In the past, simulation mannequins primarily have been used in hospital training centers to teach physicians new procedures. But by hoisting mannequins onto gurneys and sending them into patient rooms, Stanford is taking simulation in an innovative direction. “This is new ground,” Lighthall said.

Lighthall and a team of four professionals from Stanford’s Center for Immersive and Simulation-based Learning spend at least two hours preparing for each simulation exercise. They program a mannequin, which has a breathing apparatus and can generate electronic wave forms on an ECG machine, for the kind of critical event being tested—hemorrhage, allergic reaction, respiratory distress. The team then gives the nurse manager a clinical history of the patient and, in a case involving hemorrhaging, will drape bloody towels and blankets around the bed. “We say, ‘The mannequin is going to experience some problems—we can’t tell you just what, but take it seriously,’” Lighthall explained.

The critical-care specialist said initial findings from this year’s simulation exercises suggest that there is great variability in how well high-risk events are managed. The goal is to find ways of ensuring that the highest levels of performance are the rule, rather than the exception, and he thinks some improvements can be made in the availability of key sets of information.

Physicians already carry printed cards in their pockets that spell out the protocols for cardiac arrests, and similar cognitive aids could be prepared for other life-threatening events, such as how to obtain blood and manage a massive transfusion for a hemorrhaging patient. And because administering a massive transfusion requires particular skill and experience, Lighthall said, his team also envisions more focused training of designated physicians and nurses to create so-called “pockets of expertise” that could be sent to emergencies throughout the hospital.

Finally, Lighthall said, the simulation exercises show that precise communication is fundamental. For example, if a nurse calls the transfusion service and says, “We need two units of blood,” that may not sound like an emergency to a blood bank technician who is trained to listen for, “This guy is bleeding to death.” Communication in both directions, he added, “needs to be very precise and accurate, and in tune with the gravity of the situation.”

As Lighthall and his colleagues debriefed the medical team that had cared for “Mr. Overton,” nurse DeGuzman had one final question: “Did he survive?”

Thumbs up. Mr.Overton would be back to bleed another day.