The company that developed the technology made an estimate. Surgical teams, it said, would need about eight operations before they could do the new procedure as fast as a conventional one. For even the fastest-learning teams in one study, the number was closer to 40.
The procedure, introduced in hospitals in the late 1990s, was minimally invasive cardiac surgery. Instead of splitting the breastbone, the surgeon worked on the heart through a small incision between the ribs. That changed everyone’s job. A balloon threaded into the aorta now served as a clamp, and the surgeon could neither see nor feel it; the anesthesiologist guided it by ultrasound, and the nurse and the perfusionist watched that it stayed in place. Information the surgeon used to get by sight and touch now came from monitors out of his or her view. The surgeon had to rely on the team, which disrupted, as the researchers put it, “the surgeon’s role as order giver.”
The researchers, Amy Edmondson, Richard Bohmer and Gary Pisano, followed 16 teams adopting the procedure, collecting data on 660 patients and interviewing the staff involved. Every team got the same three-day training. At first, at every hospital, the operation took two to three times longer than open-heart surgery, though deaths and serious complications were no higher. All the teams got faster. The pace differed dramatically.
At a hospital the researchers called Chelsea (all the names are pseudonyms), the new chief of cardiac surgery was renowned and had already used the procedure in numerous operations elsewhere. His team was assembled by seniority, and he didn’t take part in its dry run. “It was not a matter of training myself,” he explained. “It was a matter of training the team.” That training, he said, wouldn’t require a change in how he communicated with the team: “Once I get the team set up, I never look up.” After almost 50 cases he said, “It doesn’t seem to be getting that much better.”
At Mountain Medical Center, a community hospital, the leader was a relatively junior surgeon who picked team members for their experience working together. “The ability of the surgeon to allow himself to become a partner, not a dictator, is critical,” he said. One team member said the “hierarchy [has] changed.”
Educational background and surgical experience didn’t necessarily predict how fast a team learned, the researchers found, and neither did the surgeon’s status. The fastest teams were built for learning, their leaders made the challenge motivating, and those leaders created psychological safety. When one person learns, trial and error happens in private. On a team, the authors wrote, people “risk appearing ignorant or incompetent when they suggest or try something new.” Teams whose members felt comfortable making suggestions, trying things that might not work, pointing out problems and admitting mistakes learned faster. The training had told everyone to speak up, yet such feedback often didn’t happen. One team member was upbraided for pointing out what he believed was a life-threatening situation. In their fuller academic paper, the authors reported that successful hospitals designed preparatory practice sessions and early trials to create psychological safety.
At one hospital, a nurse suggested in the middle of an operation that a long-discarded clamp, known as the “iron intern,” might solve a problem. It became part of that team’s permanent routine. And Mountain Medical, despite its leader’s modest reputation, became one of the two hospitals in the study that learned the new technology most quickly.