The Checklist Manifesto (6 page)

In 2003, however, the Michigan Health and Hospital Association approached Pronovost about testing his central line checklist throughout the state’s ICUs. It would be a huge undertaking. But Pronovost would have a chance to establish whether his checklists could really work in the wider world.

I visited Sinai-Grace Hospital, in inner-city Detroit, a few years after the project was under way, and I saw what Pronovost was up against. Occupying a campus of red brick buildings amid abandoned houses, check-cashing stores, and wig shops on the
city’s West Side, just south of Eight Mile Road, Sinai-Grace is a classic urban hospital. It employed at the time eight hundred physicians, seven hundred nurses, and two thousand other medical personnel to care for a population with the lowest median income of any city in the country. More than a quarter of a million residents were uninsured; 300,000 were on state assistance. That meant chronic financial problems. Sinai-Grace is not the most cash-strapped hospital in the city—that would be Detroit Receiving Hospital, where more than a fifth of the patients have no means of payment. But between 2000 and 2003, Sinai-Grace and eight other Detroit hospitals were forced to cut a third of their staff, and the state had to come forward with a $50 million bailout to avert their bankruptcy.

Sinai-Grace has five ICUs for adult patients and one for infants. Hassan Makki, the director of intensive care, told me what it was like there in 2004, when Pronovost and the hospital association started a series of mailings and conference calls with hospitals to introduce checklists for central lines and ventilator patients. “Morale was low,” he said. “We had lost lots of staff, and the nurses who remained weren’t sure if they were staying.” Many doctors were thinking about leaving, too. Meanwhile, the teams faced an even heavier workload because of new rules limiting how long the residents could work at a stretch. Now Pronovost was telling them to find the time to fill out some daily checklists?

Tom Piskorowski, one of the ICU physicians, told me his reaction: “Forget the paperwork. Take care of the patient.”

I accompanied a team on 7:00 a.m. rounds through one of the surgical ICUs. It had eleven patients. Four had gunshot wounds (one had been shot in the chest; one had been shot through the bowel, kidney, and liver; two had been shot through the neck and
left quadriplegic). Five patients had cerebral hemorrhaging (three were seventy-nine years and older and had been injured falling down stairs; one was a middle-aged man whose skull and left temporal lobe had been damaged by an assault with a blunt weapon; and one was a worker who had become paralyzed from the neck down after falling twenty-five feet off a ladder onto his head). There was a cancer patient recovering from surgery to remove part of his lung, and a patient who had had surgery to repair a cerebral aneurysm.

The doctors and nurses on rounds tried to proceed methodically from one room to the next but were constantly interrupted: a patient they thought they’d stabilized began hemorrhaging again; another who had been taken off the ventilator developed trouble breathing and had to be put back on the machine. It was hard to imagine that they could get their heads far enough above the daily tide of disasters to worry about the minutiae on some checklist.

Yet there they were, I discovered, filling out those pages. Mostly, it was the nurses who kept things in order. Each morning, a senior nurse walked through the unit, clipboard in hand, making sure that every patient on a ventilator had the bed propped at the right angle and had been given the right medicines and the right tests. Whenever doctors put in a central line, a nurse made sure that the central line checklist had been filled out and placed in the patient’s chart. Looking back through the hospital files, I found that they had been doing this faithfully for more than three years.

Pronovost had been canny when he started. In his first conversations with hospital administrators, he hadn’t ordered
them to use the central line checklist. Instead, he asked them simply to gather data on their own line infection rates. In early 2004, they found, the infection rates for ICU patients in Michigan hospitals were higher than the national average, and in some hospitals dramatically so. Sinai-Grace experienced more central line infections than 75 percent of American hospitals. Meanwhile, Blue Cross Blue Shield of Michigan agreed to give hospitals small bonus payments for participating in Pronovost’s program. A checklist suddenly seemed an easy and logical thing to try.

In what became known as the Keystone Initiative, each hospital assigned a project manager to roll out the checklist and participate in twice-monthly conference calls with Pronovost for troubleshooting. Pronovost also insisted that the participating hospitals assign to each unit a senior hospital executive who would visit at least once a month, hear the staff ’s complaints, and help them solve problems.

The executives were reluctant. They normally lived in meetings, worrying about strategy and bud gets. They weren’t used to venturing into patient territory and didn’t feel they belonged there. In some places, they encountered hostility, but their involvement proved crucial. In the first month, the executives discovered that chlorhexidine soap, shown to reduce line infections, was available in less than a third of the ICUs. This was a problem only an executive could solve. Within weeks, every ICU in Michigan had a supply of the soap. Teams also complained to the hospital officials that, although the checklist required patients be fully covered with a sterile drape when lines were being put in, full-size drapes were often unavailable. So the officials made sure that drapes were stocked. Then they persuaded Arrow International,
one of the largest manufacturers of central lines, to produce a new kit that had both the drape and chlorhexidine in it.

In December 2006, the Keystone Initiative published its findings in a landmark article in the
New England Journal of Medicine
. Within the first three months of the project, the central line infection rate in Michigan’s ICUs decreased by 66 percent. Most ICUs—including the ones at Sinai-Grace Hospital—cut their quarterly infection rate to zero. Michigan’s infection rates fell so low that its average ICU outperformed 90 percent of ICUs nationwide. In the Keystone Initiative’s first eighteen months, the hospitals saved an estimated $175 million in costs and more than fifteen hundred lives. The successes have been sustained for several years now—all because of a stupid little checklist.

It is tempting to think this might be an isolated success. Perhaps there is something unusual about the strategy required to prevent central line infections. After all, the central line checklist did not prevent any of the other kinds of complications that can result from sticking these foot-long plastic catheters into people’s chests—such as a collapsed lung if the needle goes in too deep or bleeding if a blood vessel gets torn. It just prevented infections. In this particular instance, yes, doctors had some trouble getting the basics right—making sure to wash their hands, put on their sterile gloves and gown, and so on—and a checklist proved dramatically valuable. But among the myriad tasks clinicians carry out for patients, maybe this is the peculiar case.

I started to wonder, though.

Around the time I learned of Pronovost’s results, I spoke to Markus Thalmann, the cardiac surgeon who had been the lead
author of the case report on the extraordinary rescue of the little girl from death by drowning. Among the many details that intrigued me about the save was the fact that it occurred not at a large cutting-edge academic medical center but at an ordinary community hospital. This one was in Klagenfurt, a small provincial Austrian town in the Alps nearest to where the girl had fallen in the pond. I asked Thalmann how the hospital had managed such a complicated rescue.

He told me he had been working in Klagenfurt for six years when the girl came in. She had not been the first person whom he and his colleagues had tried to revive from cardiac arrest after hypothermia and suffocation. His hospital received between three and five such patients a year, he estimated, mostly avalanche victims, some of them drowning victims, and a few of them people attempting suicide by taking a drug overdose and then wandering out into the snowy Alpine forests to fall unconscious. For a long time, he said, no matter how hard the hospital’s medical staff tried, they had no survivors. Most of the victims had been without a pulse and oxygen for too long when they were found. But some, he was convinced, still had a flicker of viability in them, yet he and his colleagues had always failed to sustain it.

He took a close look at the case records. Preparation, he determined, was the chief difficulty. Success required having an array of people and equipment at the ready—trauma surgeons, a cardiac anesthesiologist, a cardiothoracic surgeon, bioengineering support staff, a cardiac perfusionist, operating and critical care nurses, intensivists. Almost routinely, someone or something was missing.

He tried the usual surgical approach to remedy this—yelling at everyone to get their act together. But still they had no saves.
So he and a couple of colleagues decided to try something new. They made a checklist.

They gave the checklist to the people with the least power in the whole process—the rescue squads and the hospital telephone operator—and walked them through the details. In cases like these, the checklist said, rescue teams were to tell the hospital to prepare for possible cardiac bypass and rewarming. They were to call, when possible, even before they arrived on the scene, as the preparation time could be significant. The telephone operator would then work down a list of people to notify them to have everything set up and standing by.

With the checklist in place, the team had its first success—the rescue of the three-year-old girl. Not long afterward, Thalmann left to take a job at a hospital in Vienna. The team, however, has since had at least two other such rescues, he said. In one case, a man had been found frozen and pulseless after a suicide attempt. In another, a mother and her sixteen-year-old daughter were in an accident that sent them and their car through a guardrail, over a cliff, and into a mountain river. The mother died on impact; the daughter was trapped as the car rapidly filled with icy water. She had been in cardiac and respiratory arrest for a prolonged period of time when the rescue team arrived.

From that point onward, though, everything moved like clockwork. By the time the rescue team got to her and began CPR, the hospital had been notified. The transport team delivered her in minutes. The surgical team took her straight to the operating room and crashed her onto heart-lung bypass. One step followed right after another. And, because of the speed with which they did, she had a chance.

As the girl’s body slowly rewarmed, her heart came back. In the ICU, a mechanical ventilator, fluids, and intravenous drugs kept her going while the rest of her body recovered. The next day, the doctors were able to remove her lines and tubes. The day after that, she was sitting up in bed, ready to go home.

Four generations after the first aviation checklists went into use, a lesson is emerging: checklists seem able to defend anyone, even the experienced, against failure in many more tasks than we realized. They provide a kind of cognitive net. They catch mental flaws inherent in all of us—flaws of memory and attention and thoroughness. And because they do, they raise wide, unexpected possibilities.

But they presumably have limits, as well. So a key step is to identify which kinds of situations checklists can help with and which ones they can’t.

Two professors who study the science of complexity—Brenda Zimmerman of York University and Sholom Glouberman of the University of Toronto—have proposed a distinction
among three different kinds of problems in the world: the simple, the complicated, and the complex. Simple problems, they note, are ones like baking a cake from a mix. There is a recipe. Sometimes there are a few basic techniques to learn. But once these are mastered, following the recipe brings a high likelihood of success.

Complicated problems are ones like sending a rocket to the moon. They can sometimes be broken down into a series of simple problems. But there is no straightforward recipe. Success frequently requires multiple people, often multiple teams, and specialized expertise. Unanticipated difficulties are frequent. Timing and coordination become serious concerns.

Complex problems are ones like raising a child. Once you learn how to send a rocket to the moon, you can repeat the process with other rockets and perfect it. One rocket is like another rocket. But not so with raising a child, the professors point out. Every child is unique. Although raising one child may provide experience, it does not guarantee success with the next child. Expertise is valuable but most certainly not sufficient. Indeed, the next child may require an entirely different approach from the previous one. And this brings up another feature of complex problems: their outcomes remain highly uncertain. Yet we all know that it is possible to raise a child well. It’s complex, that’s all.

Thinking about averting plane crashes in 1935, or stopping infections of central lines in 2003, or rescuing drowning victims today, I realized that the key problem in each instance was essentially a simple one, despite the number of contributing factors. One needed only to focus attention on the rudder and elevator controls in the first case, to maintain sterility in the second, and to be prepared for cardiac bypass in the third. All were amenable,
as a result, to what engineers call “forcing functions”: relatively straightforward solutions that force the necessary behavior—solutions like checklists.